environmental inequalities in europe - who | world health … · 2019. 5. 23. · environmental...

10

20

30

40

50En

viro

nmen

tal h

ealth

ineq

ualit

ies

in E

urop

e

Environm

ental health inequalities in Europe

Recent debate on the social determinants of health has indicated that the unequal distribution of health and well-being in national populations is a major challenge for public health governance. �is is equally true for environmental health condi-tions and for exposure to environmental risk, which varies strongly by a range of sociodemographic determinants and thus causes inequalities in exposure to – and potentially in disease resulting from – environmental conditions.

Interventions tackling such environmental health inequalities need to be based on an assessment of their magnitude and on the identi�cation of population groups that are most exposed or most vulnerable to environmental risks. However, data to quantify the environmental health inequality situation are not abundant, making comprehensive assessments di�cult at both national and international levels. Following up on the commitments made by Member States at the Fifth Ministe-rial Conference on Environment and Health in Parma, Italy (2010), the WHO Regional O�ce for Europe has carried out a baseline assessment of the magnitude of environmental health inequality in the European Region based on a core set of 14 inequality indicators. �e main �ndings of the assessment report indicate that socioeconomic and demographic inequalities in risk exposure are present in all countries and need to be tackled throughout the Region. However, the report also demonstrates that each country has a speci�c portfolio of inequalities, document-ing the need for country-speci�c inequality assessments and tailored interventions on the national priorities.

World Health OrganizationRegional Office for Europe

Scherfigsvej 8, DK-2100 Copenhagen Ø, DenmarkTel.: +45 39 17 17 17. Fax: +45 39 17 18 18E-mail: [email protected] site: www.euro.who.int

The WHO European Centre for Environment and Health, Bonn Office, WHO Regional Office for Europe, coordinated the development of this report.

AbstractRecent debate on the social determinants of health has indicated that the unequal distribution of health and well-being in national populations is a major challenge for public health governance. This is equally true for environmental health conditions and for exposure to environmental risk, which varies strongly by a range of sociodemographic determinants and thus causes inequalities in exposure to – and potentially in disease resulting from – environmental conditions.Interventions tackling such environmental health inequalities need to be based on an assessment of their magnitude and on the identification of population groups that are most exposed or most vulnerable to environmental risks. However, data to quantify the environmental health inequality situation are not abundant, making comprehensive assessments difficult at both national and international levels.Following up on the commitments made by Member States at the Fifth Ministerial Conference on Environment and Health in Parma, Italy (2010), the WHO Regional Office for Europe has carried out a baseline assessment of the magnitude of environmental health inequality in the European Region based on a core set of 14 inequality indicators. The main findings of the assessment report indicate that socioeconomic and demographic inequalities in risk exposure are present in all countries and need to be tackled throughout the Region. However, the report also demonstrates that each country has a specific portfolio of inequalities, documenting the need for country-specific inequality assessments and tailored interventions on the national priorities.

KeywordsENVIRONMENTAL HEALTHENVIRONMENTAL EXPOSUREHEALTH STATUS DISPARITIESSOCIOECONOMIC FACTORSRISK FACTORSRISK ASSESSMENTEVALUATION STUDIESEUROPE

ISBN 978 92 890 0260 8

© World Health Organization 2012All rights reserved. The Regional Office for Europe of the World Health Organization welcomes requests for permission to reproduce or translate its publications, in part or in full.The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the World Health Organization concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines on maps represent approximate border lines for which there may not yet be full agreement.The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by the World Health Organization in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters.All reasonable precautions have been taken by the World Health Organization to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either express or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall the World Health Organization be liable for damages arising from its use. The views expressed by authors, editors, or expert groups do not necessarily represent the decisions or the stated policy of the World Health Organization.

Language editing: Lydia WanstallLayout: Vitali Shkaruba

Address requests about publications of the WHO Regional Office for Europe to:PublicationsWHO Regional Office for EuropeScherfigsvej 8DK-2100 Copenhagen Ø, Denmark

Alternatively, complete an online request form for documentation, health information, or for permission to quote or translate, on the Regional Office web site (http://www.euro.who.int/pubrequest).

Environmental health inequalities

in Europe

Assessment report

CONTENTSContributors ��vii

Acknowledgements �� xi

Foreword �� xiii

Executive summary ��xv

Introduction ��2Rationale for health inequality assessment and monitoring ��2Assessing environmental health inequalities in the WHO European Region ��3Rationale and overview of the project ��4Benefits of inequality assessments for action��5Overview of report ��7Methodological notes ��8Data access �� 10References�� 10

Chapter 1. The concept of environmental health inequalities �� 14A historic perspective �� 14Inequality and inequity �� 15Identification and assessment of inequalities and inequities �� 15Environmental health inequality: hazard and risk �� 16Sociodemographic factors, exposure and vulnerability �� 17The psychosocial dimension in environmental health inequality �� 17Framing the problem �� 18Conclusion �� 19References�� 20

Chapter 2. Housing-related inequalities �� 22Introduction �� 23Data and methods �� 24Restrictions and data limitations �� 25Inequalities in inadequate water supply �� 26Inequalities in lack of a flush toilet in the dwelling �� 30Inequalities in lack of a bath or shower in the dwelling �� 34Inequalities in overcrowding �� 37Inequalities in dampness in the home �� 40Inequalities in keeping the home adequately warm �� 43Inequalities in keeping the home adequately cool �� 47Conclusion on housing-related inequalities �� 48References�� 49

Chapter 3. Injury-related inequalities �� 54Introduction �� 55Data and methods �� 56Restrictions and data limitations �� 58Inequalities in work-related injuries �� 58Inequalities in fatal road traffic injuries (RTIs) �� 63Inequalities in fatal poisoning �� 69Inequalities in fatal falls �� 75Conclusion on injury-related inequalities �� 79References�� 80

Chapter 4. Environment-related inequalities ��86Introduction ��87Data and methods ��88Data restrictions and limitations ��91Inequalities in noise exposure at home ��92Inequalities in lack of access to recreational or green areas ��95Inequalities in second-hand smoke exposure at home ��101Inequalities in second-hand smoke exposure at work ��105Conclusion on environment-related inequalities ��110References��111

Chapter 5. Gaps in evidence and restrictions on assessing environmental health inequalities ��116

Introduction ��116Missing data ��116Limited stratification by sociodemographic determinants ��117Data quality ��117Consistency and comparability ��118Access to data ��119Cumulative and multiple exposures ��119Country priorities ��119Data gaps and relevance for public health��120Conclusion ��121Reference ��121

Chapter 6. Priorities for action on environmental health inequalities ��124Introduction ��124Suggested subregional priorities for action ��124Suggested priorities for national action ��125

Conclusion ��130Key messages ��130Further perspectives for action ��132Reference ��134

Annex 1. National environmental health inequality fact sheets ��136

Annex 2. Examples of national practices in analysis and presentation of environmental health inequalities ��152

Annex 3. Assessment of priority areas for national action ��180

Annex 4. Country abbreviations ��188

Annex 5. Abbreviations ��190

Environmental health inequalities in Europe vii

CONTrIbuTOrS

Editorial group and authorsGabriele Bolte, Department of Occupational and Environmental Epidemiology, Bavarian Health and Food Safety Authority, Munich, GermanyMatthias Braubach, Living Environments and Health, WHO European Centre for Environment and Health, GermanyNita Chaudhuri, American University of Paris, FranceSéverine Deguen, Department of Epidemiology and Biostatistics, EHESP School of Public Health, Rennes, FranceJon Fairburn, Institute for Environment, Sustainability and Regeneration, Staffordshire University, United KingdomIngrid Fast, Maastricht University, NetherlandsLucia Isabel Fiestas, Department of Occupational and Environmental Health, EHESP School of Public Health, FrancePaata Imnadze, National Centre for Disease Control and Public Health, GeorgiaLucie Laflamme, Department of Public Health Sciences, Karolinska Institutet, SwedenFrancesco Mitis, Violence and Injury Prevention, WHO Regional Office for Europe, DenmarkGeorge Morris, Consultant in Ecological Public Health, United KingdomDenis Zmirou-Navier, Department of Occupational and Environmental Health, EHESP School of Public Health, Rennes, France

ReviewersMaria José Carroquino, Instituto de Salud Carlos III/WHO Collaborating Centre for the Epidemiology of Environment Related Diseases, SpainHanneke Kruize, National Institute for Public Health and the Environment (RIVM), NetherlandsAnna Paldy, National Institute of Environmental Health, HungaryKieron Stanley, Environment Agency, United KingdomRebecca Steinbach, Social and Environmental Health Research, London School of Hygiene and Tropical Medicine, United Kingdom

Contributors of national fact sheets and practice examples

Fact sheetsDwellings supplied with piped water by proportion of Roma population (Hungary): Anna Paldy, National Institute of Environmental Health, HungaryAttila Juhasz and Csilla Nagy, Health Organization of Government Office of the Capital City Budapest, Hungary

Lack of flush toilet by wealth status, urban/rural residence and region (Georgia): Paata Imnadze, National Centre for Disease Control and Public Health, Georgia

Lack of bath or shower in dwelling by urban/rural population and region (Kyrgyzstan): Ainash Sharshenova, Scientific and Production Centre for Preventive Medicine, Kyrgyzstan

Overcrowding by tenure group and household income (Great Britain): George Morris, Consultant in Ecological Public Health, United KingdomBob Gilmour, School of Engineering and the Built Environment, Glasgow Caledonian University, ScotlandJake Wilson, School of Engineering and the Built Environment, Glasgow Caledonian University, Scotland

viii Contributors

Dampness in dwelling by age, income and household type (Norway): Randi Jacobsen Bertelsen and Berit Granum, Division of Environmental Medicine, Norwegian Institute of Public Health, NorwayTor Morten Normann, Statistics Norway, Norway

Lack of home heating by education and household composition (Serbia and Montenegro): Dragana Vujanovic, Office for Sustainable Development of Insufficiently Developed Areas, Serbia

Work-related injuries by sex and economic sector (Croatia): Goranka Petrović, Department of Physiology, Monitoring and Improvement of Nutrition, Croatian National Institute of Public Health, Croatia

Transport-related mortality by sex and age (Malta): Roberto DeBono, Environmental Health Policy Coordination Unit, Ministry of Health, the Elderly and Community Care, Malta

Mortality from accidental poisoning by age and urban/rural residence (Poland):Krzysztof Skotak, National Institute of Public Health – National Institute of Hygiene, Poland

Mortality from falls by age and sex (Romania): Alexandra Cucu, National Centre for Health Promotion and Evaluation, National Institute of Public Health, Romania

Exposure to traffic noise within residential areas by household income level (Netherlands): Hanneke Kruize, National Institute for Public Health and the Environment (RIVM), Netherlands

Lack of access to recreational/green space by region (Spain): Maria José Carroquino, Instituto de Salud Carlos III/WHO Collaborating Centre for the Epidemiology of Environment Related Diseases, Spain

Potential smoke exposure at home by SES (Germany): Christiane Bunge, Environmental Hygiene, Federal Environment Agency, Germany

Exposure to second-hand smoke at work by sex, age, income, and employment (Italy): Andrea Ranzi, Environmental Health Reference Centre, Regional Agency for Environmental Prevention of the Emilia Romagna (ARPA), Italy

Practice examplesEnvironmental health inequality action in France: a report on the SIGFRIED Project:Julien Caudeville and Celine Boudet, National Institute for Industrial Environment and Risks (INERIS), France

The United Kingdom sustainable development indicators: reporting on environmental health inequalities: Kieron Stanley, Environment Agency, United Kingdom

Identification of environmental health inequalities in populations living close to waste disposal sites in Italy: Andrea Ranzi, Regional Reference Centre on Environment and Health, Regional Agency for Environmental Prevention of Emilia Romagna Region, ItalyChiara Badaloni and Francesco Forastiere, Department of Epidemiology, Lazio Regional Health Service, ItalyGiuseppe Costa, Department of Biological and Clinical Sciences, University of Turin, ItalyMarco Martuzzi, Health Impact Assessment, WHO European Centre for Environment and HealthFrancesco Mitis, Violence and Injury Prevention, WHO Regional Office for Europe, Denmark

Assessing and reporting on environmental health inequalities related to lack of heated space and indoor pollution in Serbia and Montenegro: Dragana Vujanovic, Office for Sustainable Development of Insufficiently Developed Areas, Serbia

Analysis and presentation of environmental health inequalities concerning “green space” in Scotland: George Morris, Consultant in Ecological Public Health, United KingdomJon Fairburn, Institute for Environment, Sustainability and Regeneration, Staffordshire University, United Kingdom

Environmental health inequalities in Europe ix

Assessment of environmental health inequalities in Finland: residential exposure to ambient air pollution from wood combustion and traffic: Pauliina Taimisto, Marko Tainio, Jouni T. Tuomisto and Matti Jantunen, National Institute for Health and Welfare (THL), FinlandNiko Karvosenoja, Kaarle Kupiainen and Petri Porvari, Finnish Environment Institute (SYKE), FinlandAri Karppinen and Leena Kangas and Jaakko Kukkonen, Finnish Meteorological Institute (FMI), Finland

Environmental health inequality report on the impact of socioeconomic status on the prevalence of allergies and respiratory diseases and symptoms in Hungarian children: Anna Paldy, Peter Rudnai, Mihaly Janos Varro, Annamaria Macsik and Eszter Szabo, National Institute of Environmental Health, Hungary Attila Juhasz and Csilla Nagy, Health Organization of Government Office of the Capital City Budapest, Hungary

Participants of project-related WHO meetings on environmental health inequalityRandi Jacobsen Bertelsen, Division of Environmental Medicine, Norwegian Institute of Public Health, NorwayGabriele Bolte, Department of Occupational and Environmental Epidemiology, Bavarian Health and Food Safety Authority, Munich, GermanyCeline Boudet, National Institute for Industrial and Environment and Risks (INERIS), FranceChristiane Bunge, Environmental Hygiene, Federal Environment Agency, GermanyKrunoslav Capak, Environmental Health Ecology Service, National Institute of Public Health, CroatiaMaria José Carroquino, Instituto de Salud Carlos III/WHO Collaborating Centre for the Epidemiology of Environment Related Diseases, SpainJulien Caudeville, National Institute for Industrial Environment and Risks (INERIS), FranceGiuseppe Costa, Department of Biological and Clinical Sciences, University of Turin, Italy Alexandra Cucu, National Centre for Health Promotion and Evaluation, National Institute of Public Health, RomaniaAryuna Dashitsyrenova, Ministry of Health and Social Development, Russian FederationRoberto DeBono, Environmental Health Policy Coordination Unit, Ministry of Health, the Elderly and Community Care, MaltaSéverine Deguen, Department of Epidemiology and Biostatistics, EHESP School of Public Health, Rennes, FranceMartin Devine, Health Services Executive, IrelandNita Chaudhuri, American University of Paris, FranceJon Fairburn, Institute for Environment, Sustainability and Regeneration, Staffordshire University, United KingdomBerit Granum, Division of Environmental Medicine, Norwegian Institute of Public Health, NorwayPaata Imnadze, National Centre for Disease Control and Public Health, GeorgiaMatti Jantunen, Department of Environmental Health, National Institute of Health and Welfare, FinlandHanneke Kruize, National Institute for Public Health and the Environment (RIVM), NetherlandsLucie Laflamme, Department of Public Health Sciences, Karolinska Institutet, SwedenGeorge Morris, Consultant in Ecological Public Health, United KingdomAnna Paldy, National Institute of Environmental Health, HungaryRifat Pamuk, Ministry of Health, TurkeyGoranka Petrović, Department of Physiology, Monitoring and Improvement of Nutrition, Croatian National Institute of Public Health, CroatiaAndrea Ranzi, Regional Reference Centre on Environment and Health, Regional Agency for Environmental Prevention of Emilia Romagna Region, ItalyAnne Reneflot, Norwegian Institute of Public Health, NorwayAinash Sharshenova, Scientific and Production Centre for Preventive Medicine, KyrgyzstanKrzysztof Skotak, National Institute of Public Health – National Institute of Hygiene, PolandKieron Stanley, Environment Agency, United KingdomRebecca Steinbach, Social and Environmental Health Research, London School of Hygiene and Tropical Medicine, United Kingdom

x Contributors

Dragana Vujanovic, Office for Sustainable Development of Insufficiently Developed Areas, SerbiaDenis Zmirou-Navier, Department of Occupational and Environmental Health, EHESP School of Public Health, Rennes, FranceIngrid Zurlyté, Centre for Health Education and Disease Prevention, Lithuania

Meeting observersCharles Price, Directorate General for Health and Consumer Protection, European Commission, LuxembourgJames Wilson, School of Built and Natural Environment, Glasgow Caledonian University, Scotland

WHO secretariatMatthias Braubach, Living Environments and Health, WHO European Centre for Environment and Health1

Andrey Egorov, Environment and Health Surveillance, WHO European Centre for Environment and HealthIngrid Fast, Student intern, WHO European Centre for Environment and HealthStefanie Fleischmann, Student intern, WHO European Centre for Environment and HealthJohanna Hanefeld, Social Determinants of Health, WHO European Office for Investment for Health and DevelopmentRokho Kim, Occupational Health, WHO European Centre for Environment and HealthMichal Krzyzanowski, Head of Office, WHO European Centre for Environment and HealthMarco Martuzzi, Health Impact Assessment, WHO European Centre for Environment and HealthSrdan Matic, Coordinator Environment and Health, WHO Regional Office for EuropeGeraldine McWeeney, Environment and Health, WHO Country Office, SerbiaFrancesco Mitis, Violence and Injury Prevention, WHO Regional Office for EuropeAngelika Nöcker, Programme Assistant, WHO European Centre for Environment and HealthDeepika Sachdeva, Programme Assistant, WHO European Centre for Environment and HealthSteffen Uffenorde, Student intern, WHO European Centre for Environment and HealthTanja Wolf, Climate Change and Health, WHO European Centre for Environment and Health

1 Project leader�

Environmental health inequalities in Europe xi

ACKNOwlEdGEmENTS

The project and the two related expert meetings were supported by funds generously provided to WHO by the German Government through the Federal Ministry for the Environment, Nature Conservation and Nuclear Safety.

Environmental health inequalities in Europe xiii

FOrEwOrd

In 2008, the final report of the WHO Commission on Social Determinants of Health2 concluded that inequalities in health are a major challenge for both development and overall progress in countries. Such inequalities also exist within environmental health; almost all countries have some groups of their population at greater risk of experiencing harmful environmental conditions than others. Socioeconomic variables such as income, employment or occupation and education are found to be especially strong determinants of environmental health risks. Demographic variables such as age, sex and ethnicity can also affect risk, and in addition can modify the relationship between socioeconomic status, environment and health. The Member States of the WHO European Region declared their commitment to act on socioeconomic and gender inequalities in the human environment and health at the Fifth Ministerial Conference on Environment and Health in Parma, Italy, in March 2010.3

This assessment report indicates that environmental health inequalities exist in all subregions and in all countries of the WHO European Region, even though countries may have different patterns of exposure and risk. The report also confirms the expectation that often, although not exclusively, exposure to environmental risks is more frequently suffered by disadvantaged population groups.

The report shows that more and better data on the distribution of environmental risks within the population of the WHO European Region are needed. For many environmental health inequalities covered in this report, data are only available for about half the countries. The assessment of environmental inequalities is further restricted by a frequent lack of data on population subgroups defined by various categories of socioeconomic or demographic variable. This report must therefore be considered an initial baseline assessment using data available from international databases. Clearly, more work is needed to provide more and better data, enabling more insightful assessments.

The existence of significant unjust and avoidable inequalities in environmental risks within a country is not acceptable, and evidence of such inequalities, as presented in this report, thus calls for relevant policies and interventions. In consequence, the environmental health inequalities identified in the respective countries need to be validated and interpreted in the given national context, allowing the design of intersectoral remedial actions4 as well as the integration of health equity considerations into all national policies.5 Such interventions would prove that Member States have not only the capacity to identify inequalities in environmental risk but also the political will to address these inequalities and provide environmental justice as declared by the Member States in Parma.

Zsuzsanna JakabWHO Regional Director for Europe

2 Commission on Social Determinants of Health (2008)� Closing the gap in a generation: health equity through action on the social determinants of health. Final report of the Commission on Social Determinants of Health� Geneva, World Health Organization (http://whqlibdoc�who�int/publications/2008/9789241563703_eng�pdf, accessed 11 January 2012)�

3 WHO (2010)� Parma Declaration on Environment and Health. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/ __data/assets/pdf_file/0011/78608/E93618.pdf, accessed 11 January 2012).

4 WHO (2011)� Rio Political Declaration on Social Determinants of Health� Geneva, World Health Organization (http://www�who�int/sdhconference/declaration/Rio_political_declaration�pdf, accessed 11 January 2012)�

5 WHO (2009)� World Health Assembly resolution 62.14. Reducing health inequities through action on the social determinants of health� Geneva, World Health Organization (http://apps.who.int/gb/ebwha/pdf_files/A62/A62_R14-en.pdf, accessed 11 January 2012).

Environmental health inequalities in Europe xv

ExECuTIvE SummAry

THE CONCEPT OF ENvIrONmENTAl HEAlTH INEquAlITIESEnvironmental health inequalities refer to general differences in environmental health conditions. Socioeconomic and demographic inequalities in exposure to environmental hazards exist everywhere and can be expressed in relation to factors that may affect the risk of being exposed, such as income, education, employment, age, sex, race/ethnicity and specific locations or settings. In addition to these differences in exposure, environmental health inequalities are also caused by social or demographic differences in vulnerability towards certain risks.

Many of the environmental health inequalities, particularly where they are linked to socioeconomic variables or sex, also represent “inequities” because they are unfair, unjust and avoidable. The root cause of such inequalities is most often a lack of “distributive justice”, indicating that environmental risks are not evenly distributed within societies and populations, and a lack of “procedural justice”, indicating that different population groups may have different opportunities to influence decisions affecting their close environment.

rATIONAlE OF THE rEPOrTThe objective of the report is to provide an initial baseline assessment of environmental health inequalities in the WHO European Region. It is based on available statistical data from national or international databases. To undertake the assessment, a set of 14 environmental health inequality indicators was developed, categorized into three inequality dimensions (see Table).

Table. Environmental health inequality indicators

Housing-related inequalities Injury-related inequalities Environment-related inequalities

• Inadequate water supply • Lack of a flush toilet • Lack of a bath or shower • Overcrowding • Dampness in the home • Inability to keep the home

adequately warm

• Work-related injuries • Fatal road traffic injuries • Fatal poisonings • Fatal falls

• Noise exposure at home • Lack of access to green/

recreational areas • Second-hand smoke exposure

at home • Second-hand smoke exposure

at work

For each environmental health inequality indicator, data from international databases were analysed to assess, by country or subregion, the existence and the magnitude of inequalities between different population subgroups.

National data were analysed for the development of national environmental health inequality fact sheets and practice examples (see Annexes 1 and 2). These national contributions indicate that more detailed assessments of environmental health inequality can be provided at national and subnational levels and that there are already national experiences with such assessments.

INEquAlITy ASSESSmENT FINdINGSThe assessment of housing-, injury- and environment-related inequalities shows that inequalities exist throughout the WHO European Region. However, there are large differences between countries regarding the magnitude of the inequalities and the most affected population groups. Depending on

xvi Executive summary

the available data, inequality assessments were undertaken in relation to differences by sex, age, income, relative poverty, household type, social position, employment, occupation, education and difficulty paying bills. All of these sociodemographic determinants are found to be associated with significant inequalities.• Income and poverty-related inequalities are identified for noise exposure, exposure to second-hand

tobacco smoke at home and at work, and housing-related inequality indicators, where they are most clearly expressed. Compared to the other determinants applied, income- and poverty-related determinants display some of the strongest inequalities at subregional and national levels. Differences in national income levels are also associated with injury-related fatalities, with low/middle income countries reporting higher mortality rates.

• Sex-related inequality is most strongly associated with injury, where male fatality rates are often three times (and beyond) female fatality rates. Sex-related differences also appear in relation to second-hand tobacco smoke exposure, yet play no important role for housing-related risk factors.

• Age-related inequalities are present for injuries (especially falls) but differ in direction, depending on the indicator. Age impacts are less prominent for the other inequality indicators.

• Household type-related inequalities in housing conditions are especially identified for single-parent households, and increase when combined with low income and relative poverty factors.

• Data on inequalities by education, employment/occupation and self-assessed social position are only available for some of the environment-related inequalities, but they show a diverse inequality pattern: high education level is consistently associated with higher reported lack of access to recreational and green areas, while employment/occupation level shows different inequality patterns in exposure to second-hand smoke, with the direction of inequality depending on sex and subregion.

SuGGESTEd PrIOrITIES FOr NATIONAl ACTIONSuggested priorities for national action are identified in the report, based on a combined assessment of the absolute magnitude of the respective environmental exposure for the whole population and the relative exposure differences between selected population subgroups. If the respective environmental health risk is greater in one country than in others, and if the distribution of the risk within the population is more unequal in that country than in others, the country thus identified should give priority to national follow-up activities in order to address these inequalities.

Suggested priorities for national action on inequalities are identified for 38 of the 53 countries of the WHO European Region and affect Member States from all subregions and developmental levels. However, of the 15 countries where no priority for national action on environmental health inequalities was identified, 12 countries only reported data for 5 or even fewer of the 30 assessed inequality dimensions covered within the 14 environmental health inequality indicators.

Annex 3 shows the suggested priorities for the individual countries of the WHO European Region. In countries with identified priorities for national action, a more detailed national assessment of the respective inequalities is needed in order to confirm and interpret them in the given national context. However, in countries where no data were available, this lack of information should, in and of itself, be a reason for more detailed investigation.

CONSTrAINTS ANd EvIdENCE GAPSThe assessment report is affected by a range of constraints and gaps in evidence. The most significant constraints are (a) the lack of general data on environmental exposure in many countries, and (b) the limited opportunities for stratification of environmental exposure data by socioeconomic or demographic determinants. Further constraints relate to the quality and reliability of the data, and the lack of

Environmental health inequalities in Europe xvii

methodological consistency between national surveys, restricting the comparison of data collected in different countries. Priority steps to be taken towards the improvement of statistical evidence for environmental health inequality assessments would comprise:• establishment of surveys covering priority environmental health issues and specific target groups• increased use of social and demographic variables in environmental surveys• development of common tools, methods, definitions and criteria• better access to the available data.

CONCluSIONThe report conveys four key messages.• Environmental health inequalities exist in all subregions and in all countries, and are most often

suffered by disadvantaged population groups. • The magnitude of inequalities and the distribution of inequalities between advantaged and

disadvantaged population groups can be very diverse between countries and also depends on the socioeconomic or demographic variable used for stratification.

• To allow reliable identification of the most relevant target groups and to understand better the national inequality patterns and their causal mechanisms, more detailed environmental health inequality reporting and assessment are needed at the national level.

• The evidence base for the assessment of environmental health inequalities needs to be strengthened. This is valid for both data quantity (number of countries with data, number of risk factors reported) and data quality (reliability, opportunities for stratification).

Therefore, the results presented in this report provide an initial baseline assessment of selected environmental health inequalities in the WHO European Region. Further work is necessary to expand and further refine the assessment.

POSSIblE ACTIONS FOr TACKlING ENvIrONmENTAl HEAlTH INEquAlITIESAlthough national priorities and disadvantaged groups vary, action is necessary throughout the WHO European Region to reduce the observed inequalities. The report suggests six general recommendations for action, which can be tailored to the respective national situation:• action 1: general improvement of environmental conditions, assuring healthy environments for all;• action 2: mitigation and reduction of risk exposure in the most affected population groups, focusing

on the most exposed and/or most vulnerable subpopulations;• action 3: national environmental health inequality assessments to assess or confirm inequalities

based on national, more detailed data;• action 4: sharing experiences and case studies on successful interventions tackling environmental

health inequalities;• action 5: review and modification of national intersectoral policies in relation to environmental

health inequalities;• action 6: monitoring of environmental health inequalities using a standard set of inequality indicators.

INTRODUCTION

Introduction2

INTrOduCTIONMatthias Braubach

“Decades of experience tell us that this world will not become a fair place for health all by itself.” (Margaret Chan, foreword to Blas and Sivasankara Kurup, 2010)

The Constitution of the World Health Organization, established in 1946, provides WHO with a mandate to strive for the highest possible level of health for all people, irrespective of their social status, ethnicity, sex or age (WHO, 1946). The 1978 Declaration of Alma-Ata confirms this priority, defining health as a fundamental human right and stating that “inequality in the health status of the people particularly between developed and developing countries as well as within countries is politically, socially and economically unacceptable and is, therefore, of common concern to all countries” (WHO, 1978). However, the Declaration of Alma-Ata also points out that action cannot rest within the health sector alone if the goal of attaining health for all is to be met, as health is achieved as a result of the policies and actions of many sectors. This is even more evident when looking at environmental health issues: because these issues are heavily influenced by the way we live, travel, work and consume, they cannot directly be affected or even mitigated by the health sector. Therefore, the provision of adequate and equal conditions – through environmental, social and infrastructural measures – is a task for all sectors and calls for intersectoral action (WHO, 2011a) and a “health in all policies” (HiAP) approach (Ministry of Social Affairs and Health, Finland, 2006).

rATIONAlE FOr HEAlTH INEquAlITy ASSESSmENT ANd mONITOrINGThe final report of the WHO Commission on Social Determinants of Health (CSDH), Closing the gap in a generation (CSDH, 2008), shows that inequalities in health are a major challenge for both development and overall progress in countries. The report provides strong evidence showing that the true causes of health inequalities reside in the social, economic and political environments shaping the conditions in which people live. These environments are affected by laws and regulations and can therefore be improved to ensure greater fairness and equality by developing new or modifying existing policies.

However, the way events unfold in reality can be quite different. Within almost all countries some groups of the population are at greater risk of experiencing harmful environmental conditions as a result of their sociodemographic circumstances. This environmental dimension of inequality and its multiple facets – known as environmental justice or environmental (in)equality – has in recent years been increasingly recognized and documented by both researchers and national governments.

There are significant sociodemographic inequalities in both exposure to and negative health outcomes arising from adverse environmental conditions. Such inequalities exist between countries, within countries and within communities and can devided into two categories – socioeconomic and demographic inequalities. Socioeconomic status (SES) variables such as income, employment, occupation and education are found to be especially strong determinants of environmental health risks. Demographic variables such as age, gender and ethnicity can modify the relationship between SES, environment and health, and can also directly affect exposure and health-related inequalities arising from biological, social, cultural and behavioural differences.

Environmental health inequalities in Europe 3

Calling on governments to close the gap in a generation, CSDH (2008) recommended three principles of action.• Improve daily living conditions.• Tackle the inequitable distribution of power, money and resources.• Measure and understand the problem and assess the impact of action.

The first recommendation is strongly related to the environmental conditions to which people are exposed on a daily basis. The third recommendation calls for assessment of the health situation in Member States in order to understand the problem, identify causal mechanisms and set priorities for action. Most importantly, it calls for monitoring of the impacts of actions and interventions. In the context of environmental health inequalities, various reporting and monitoring opportunities arise to connect environmental exposure data with sociodemographic information to describe the disparities of environmental risk within and between population groups.

In response to the CSDH final report, the World Health Assembly agreed in 2009 on a resolution to reduce health inequities through action on the social determinants of health (WHO, 2009a). This resolution provides WHO with a strong mandate to address the social determinants of health in its work, and urges Member States to:• tackle health inequities within as well as across countries;• develop mechanisms to integrate inequalities into public health actions; • consider inequity arguments in their policy-making.

The need for monitoring of and action on inequalities was confirmed by the World Conference on Social Determinants of Health held in Brazil in October 2011. The conference discussion paper (WHO, 2011b) indicates that a dearth of knowledge – resulting from an absence of inequality monitoring and of political accountability – is one of the main reasons for the lack of action and thus calls for increased reporting of inequalities using disaggregated data rather than information based on national averages.

ASSESSING ENvIrONmENTAl HEAlTH INEquAlITIES IN THE wHO EurOPEAN rEGIONIn 2009 the WHO Regional Office for Europe began to review the evidence on social and gender inequalities in environmental risk and exposure, drafting a policy brief and an evidence report on environmental inequalities, published at the Fifth Ministerial Conference on Environment and Health in 2010 (WHO, 2010a; 2010b). At the Conference Member States recognized environmental health inequalities as a priority for future work and adopted the Parma Declaration (WHO, 2010c), which provides WHO with a mandate to monitor the commitment of Member States to act on:• the health risks to children and other vulnerable population groups (with a specific focus on the

water and sanitation situation);• socioeconomic and gender inequalities in the human environment that are relevant for health.

Work undertaken by academic researchers, as well as by WHO and other international agencies, indicates that countries already have a significant ammount of information on the most vulnerable groups in relation to specific environmental threats. However, the respective data are often scattered and rarely brought together in a systematic way, as concluded by the first WHO expert meeting to review the evidence on environmental inequalities in 2009 (WHO, 2009b). Following up on this lack of quantitative evidence, and based on the Parma Declaration and the World Health Assembly resolution, WHO identified the need to assess in more detail the environmental health inequalities in the WHO European Region as a basis for further action by WHO and Member States.

Introduction4

rATIONAlE ANd OvErvIEw OF THE PrOjECTThe main objective of the project was to assess and report on environmental health inequalities in the WHO European Region, based on available statistical data from national or international databases. The results of the project are presented in this assessment report, which describes the magnitude of environmental disparities in the WHO European Region and identifies the population groups that are most affected. Some academic research work referenced in the report chapters is also applied to put the statistical data into context, to provide information about the health relevance of the risk factors covered, and to shed light on inequalities that cannot be assessed through the data alone. However, emphasis is also placed on identifying and reporting the gaps in evidence that restrict the assessment of environmental inequalities, which may be as relevant for public health practitioners and policy-makers as the findings.

Initiated in 2010 and benefitting from the work reviewing academic evidence for the Fifth Ministerial Conference on Environment and Health, the project aimed to:• establish a list of environmental health risk factors for which data on socioeconomic or demographic

inequalities can be compiled at the national level;• select and apply an environmental health inequality indicator set for the assessment of country-

specific environmental health inequality data;• produce an assessment report on environmental health inequality together with national inequality

fact sheets.

Two milestone meetings took place at the WHO European Centre on Environment and Health in Bonn to mark the project’s progress.

Identification of available data and selection of inequality indicatorsInformed by the evidence review published for the Fifth Ministerial Conference on Environment and Health (WHO, 2010b), a compilation of available data on environmental health risk factors and their potential for stratification by sociodemographic determinants was put together during summer 2010. 17 Member States participated, searching for available data based on national censuses and surveys. In parallel, the WHO secretariat reviewed international databases (including those of the European Union (EU), the Organisation for Economic Co-operation and Development (OECD) and the United Nations) for data on the environmental health risk factors that could be stratified by socioeconomic or demographic determinants.

In October 2010 at the first project meeting 26 experts from different countries across the WHO European Region and WHO staff from various programmes reviewed and evaluated the compilation of environmental health inequality data and data sources (WHO, 2010d). From an initial 30 risk factors compiled by the WHO secretariat and the Member States, a set of 14 environmental health inequality indicators categorized into three inequality areas (housing-, injury- and environment-related inequalities) was agreed (see Table 1).

Although the review was undertaken on both the national and international levels, the meeting concluded that the data compiled through national surveys and censuses were too diverse to be used for international reporting. The main restrictions relate to the variety of collection methods and definitions used, which do not enable consistent and reliable comparison between countries (see Chapter 5 for details). Nevertheless, national data have strong potential to be useful in environmental health inequality assessments within individual countries, as indicated by the national inequality fact sheets and practice examples in Annexes 1 and 2.


Table 1. Environmental health inequality indicators

Indicator Sociodemographic stratification options available data source

Housing-related inequalities

Inadequate water supply Urbanization level WHO/UNICEF

Lack of a flush toilet Age, sex, income/poverty status and household type Eurostat

Lack of a bath or shower Age, sex, income/poverty status and household type Eurostat

Overcrowding Age, sex, income/poverty status and household type Eurostat

Dampness in the home Age, sex, income/poverty status and household type Eurostat

Inability to keep the home adequately warm Age, sex, income/poverty status and household type Eurostat

Injury-related inequalities

Work-related injuries Sex, age and occupation Eurostat

Fatal road traffic injuries Country income, age and sex WHO

Fatal poisonings Country income, age and sex WHO

Fatal falls Country income, age and sex WHO

Environment-related inequalities

Noise exposure at home Income/poverty status and household type Eurostat

Lack of access to green/recreational areas Age, sex, income, difficulty paying bills, employment, education level and household type

Eurofound

Second-hand smoke exposure at home Age, sex, self-assessed social position, difficulty paying bills and employment

Eurobaro meter

Second-hand smoke exposure at work Age, sex, self-assessed social position, difficulty paying bills and occupation

Eurobaro meter

Implementation of indicators and drafting the first assessment reportUsing the environmental health inequality indicators, subcontracted experts drafted reports assessing indicator-specific inequalities. Chapters on the concept of environmental health inequality and on the gaps in evidence identified as restricting a more detailed inequality assessment were developed in parallel. In addition, selected Member States were asked to provide fact sheets on environmental health inequalities (showing the potential for more detailed assessments using national data) and practice examples describing the experience and methods applied in recent national work identifying, monitoring and assessing environmental health inequalities.

At the second project meeting in June 2011, the chapter drafts, fact sheets and practice examples were peer-reviewed and discussed by 24 country representatives and experts. The chapters were then revised and finalized, based on their comments.

bENEFITS OF INEquAlITy ASSESSmENTS FOr ACTIONAction to tackle inequalities needs to be informed by evidence on the population groups most affected and the sociodemographic features associated with the unequal distribution of risks and opportunities. Hence, better quality evidence and adequate identification of the specific target groups could help to make interventions more effective. Table 2 indicates the potential benefits of using inequality evidence for policy action, suggesting that such actions can be focused on societal structures and mechanisms as well as on resulting disparities in exposure and/or vulnerability.

Introduction6

Table 2. Benefits of inequality reporting for effective action on environmental health inequalities

Inequality evidence Policy actions

Evidence on societal structures and mechanisms leading to inequalities

• Provide examples of good/equitable societal practices� • Review and propose policy options on environmental protection�• Engage in public debate to incorporate health equity issues into economic and social

strategies and plans�• Support and implement equity-focused health impact assessment of policies and

infrastructural projects�Evidence on differential exposure to social and physical environmental risks

• Advocate for appropriate interventions to improve environmental conditions for the whole population�

• Target action on pollution hotspots and population groups with the highest exposures�• Influence the health ministry to shift attention upstream to policies that produce good

population health�• Support intersectoral action and extend HiAP approaches�• Actively participate in public education, regulation, infrastructure planning and design,

and taxation policy development affecting environmental conditions�Evidence on differential vulnerability to the risks

• Ensure adequate environmental and infrastructural services and conditions throughout each country�

• Increase targeted protection measures in areas or settings with a high density of vulnerable, sensitive or disproportionally affected populations�

• Improve environmental standards in the vicinity of child care centres, schools, hospitals, nursing homes, and similar�

Source: adapted and extended, based on a concept outlined in Blas, Sommerfeld and Sivasankara Kurup (2011).

Since the transformation of society structures and procedures may be more of a long-term objective for the improvement of health for all, the reduction of environmental inequalities specifically requires short-term interventions in decreasing exposure (Braubach et al., 2010). In this context, Table 2 indicates that in many cases the decision will be between two separate approaches: interventions assuring environmental conditions for all and targeted interventions tackling environmental conditions specific to certain groups or geographical units. Although both approaches are needed and can often be combined to achieve the best outcome (Dahlgren and Whitehead, 2006), the results of inequality assessments are essential to inform the decision-making process and provide guidance on the most appropriate way forward.

On the one hand, as indicated by the recent report on equity, social determinants and public health programmes (Blas and Sivasankara Kurup, 2010), the existence of linear gradients of environmental inequality (see Fig. 1, Country A) would strongly suggest that universal approaches – improving environmental conditions to reduce exposure for all groups, irrespective of social status – would be beneficial for the whole of society. Environmental actions such as assuring compliance with existing environmental standards throughout the country are likely to have the greatest benefits for the most disadvantaged segment of the population with the highest levels of exposure. Thus, broad-brush environmental actions might help to reduce inequalities more effectively.

On the other hand, actions should not ignore the specific needs of population groups with higher social disadvantage, which might benefit most from dedicated action. Thus, environmental inequality gradients with a skewed distribution (see Fig. 1, Country B) – especially those with excess risk for the poorest population groups – would benefit from the application of targeted actions focusing on the environmental conditions suffered by the most disadvantaged population groups, which, in addition, are often less socially included and less involved in political advocacy.

This assessment report provides examples that show how analysis based on the results of statistical data can assist with the selection of appropriate interventions. However, this approach requires the identification of gradients which cannot be produced when only dichotomous comparisons of, for example, “rich versus poor” and “male versus female” are possible. As demonstrated by the results in this report, data availability often does not facilitate adequate assessment of inequalities using these existing gradients.


Fig. 1. Examples of linear and nonlinear inequality gradients

0

5

10

15

20

25

30

1 (strong socialdisadvantage)

2 3 4 5 (no socialdisadvantage)

Exp

osur

e pr

eval

ence

(%)

Country ACountry B

OvErvIEw OF rEPOrTThe assessment report begins with an introduction to the concept of environmental health inequalities (Chapter 1). It presents the historic development of the approach and provides insight into the terminology and scientific concepts used. This introduction will provide readers unfamiliar with the concept of environmental health inequality with a basic understanding of the field.

The conceptual introduction is followed by the three main chapters of the assessment report, presenting the environmental health inequalities in the WHO European Region. Chapter 2 looks at housing-related inequalities, covering those related to water and sanitation (water supply and sanitary equipment within homes) as well as those related to the quality and size of the dwelling (overcrowding, dampness and thermal comfort). This is followed by Chapter 3 on injury-related inequalities, which assesses the unequal distribution of work-related injuries and fatal traffic injuries, poisonings and falls. Chapter 4 then considers the inequalities in noise exposure, access to green and recreational areas, and second-hand smoke exposure at home and at work. Each of these assessment chapters also includes a section on the health relevance of the identified inequalities as well as main conclusions and suggested mitigation actions.

The presentation of the findings is complemented by a review of the evidence gaps and the barriers to assessing environmental health inequalities. Chapter 5 shows that the findings presented are far from exhaustive and argues that there are still fundamental gaps in the evidence yet to be tackled.

Chapter 6 then merges the findings to identify the patterns of and possible priority areas for action on environmental health inequality observed in the WHO European Region. The assessment results are reviewed to highlight:• the main inequalities found in the European subregions; • the countries facing the largest challenges of environmental health inequalities based on a

combination of the absolute magnitude of an environmental problem (prevalence levels or mortality rates in the total population) and the magnitude of relative inequality between selected population subgroups.

Finally, the report concludes by summarizing the most relevant key messages, and provides six recommendations for potential action.

Introduction8

Complementing and extending the international assessment report and its findings, three annex sections focus on evidence, experiences and suggested priorities for action at the national level. Annex 1 presents a national fact sheet for each of the 14 environmental health inequality indicators and shows that proper assessment of environmental health inequalities can be undertaken in all Member States of the WHO European Region, irrespective of social or economic level. Annex 2 provides specific examples and experiences from selected Member States, showing steps and methods that can be applied to identify and assess environmental health inequalities. Annex 3 presents in detail the assessment of suggested priorities for national action on range of environmental health inequality dimensions, as presented in Chapter 6.

mETHOdOlOGICAl NOTESAs mentioned above, availability and consistency of data were the main challenges for putting together this assessment report. The variety of data formats and the available stratification options by sociodemographic determinants have also had an impact on the chapter contents. Table 1 above shows that for each of the three inequality dimensions (housing, injury and environment) different data sources were used for the assessment of environmental health inequalities. Rather than applying one common methodological approach to all indicators, the authors have tried to adapt the analysis to the available information by choosing the most practical analysis methods for the respective data. As a result, each chapter has approached the assessment and presentation of the inequality situation slightly differently. Chapter-specific sections on the data and methods used inform the reader about the information available for the assessment, the associated constraints, and the methods applied.

Nevertheless, throughout the assessment report, the authors have attempted to present two different inequality dimensions:• absolute dimensions of inequality, as shown by absolute differences in, for example, mortality rates

or environmental exposure prevalence levels between population groups; • relative dimensions of inequality (where appropriate), as shown by ratios comparing, for example,

the excess mortality in or prevalence of the most affected population group to the less affected or the total population, thus showing the relative magnitude of inequality.

While the authors believe that a complete assessment of environmental health inequalities must be based on both absolute and relative inequality dimensions, the data sometimes made this approach difficult in practical terms. Several examples shown in this assessment report indicate that the highest relative inequalities can often be found in countries where the overall prevalence of a given problem is very low. Therefore, it is necessary to note that any relative expression of inequality always needs to be interpreted in light of the overall prevalence situation, as well as the absolute differences between the compared population groups. For example, if the overall prevalence of an environmental problem in the general population is 1%, the lowest-income subgroups might have a prevalence of 5%, while the highest-income subgroups might have a prevalence of only 0.5%. The relative difference between these income groups is then described by a ratio of 10:1, while the absolute difference is 4.5%. In comparison, countries with an overall prevalence of 10% in the general population rarely achieve such high relative inequality ratios. It should also be noted that the same absolute difference of 4.5% provides a ratio of only 1.45:1 if the population groups compared have prevalence levels of 14.5% and 10%. Bearing this in mind, it is clear that in countries where both the prevalence of an environmental problem and the relative contrast in prevalence between subgroups are high, political action is more urgently required.

The sample sizes of the surveys used as data sources presented another constraint. Many of these surveys (such as those coordinated by Eurostat or Eurofound) are designed to provide nationally representative estimates for a range of variables for the total population of the country. However, when analysing such datasets from an inequality perspective, the data are divided into population subgroups, quickly reducing the respective sample size. For example, in the United Kingdom the sample size of households


participating in the Eurostat survey on EU Statistics on Income and Living Conditions (EU-SILC) is 7500. Analysing the effect, for example, of poverty (which may affect only 15% of the population) in single-parent households (which may be 5% of all households) reduces this sample size to only 50–60 households that represent single-parent households living in poverty. Similarly, the size of the respective sample is affected by the prevalence level of a given environmental problem, with lower prevalence levels reducing the sample size further. This makes the results less reliable and the findings less representative of the general population.

Another factor was the population size of the respective country: compared to the United Kingdom, which includes 7500 households in the EU-SILC survey, smaller countries such as Ireland and Estonia provide even smaller samples to start with (3750 and 3500 households respectively). Other surveys used, such as the Eurobarometer and European Quality of Life Survey (EQLS), are based on even lower sample sizes. Therefore, the assessment of environmental inequalities between population subgroups may suffer from poor reliability. Nevertheless, these databases seem to be the only sources providing consistent and comparable data for assessment of environmental health inequalities.

A further challenge was that many of the international databases are frequently updated. The main work on the assessment report was undertaken in 2011, all data having been downloaded in spring 2011. Final modifications and changes were made in late 2011 when data for 2010 started to become available for some (but not the majority of ) countries. Therefore, the inequality assessment is based on data reported for 2009 or, where this was not available, the last year of reporting for the respective countries. However, the lack of data for many countries, especially non-EU countries, is of much greater concern.

In order not only to compare countries but also to assess the inequality conditions by geopolitical subregion, the data were aggregated to reflect four subregions of the WHO European Region. Subregional categorization reflects the geographical and political situation as indicated by Table 3 and Map 1.5

Table 3. European subregions used for the assessment

Subregion Country coverage

Euro 1 (21 countries) All countries belonging to the EU before May 2004 and western European countries on comparable developmental level (such as Norway and Switzerland)

Euro 2 (12 countries) All countries joining the EU after May 2004Euro 3 (12 countries) All countries belonging to the former Soviet Union (except the Baltic states)Euro 4 (8 countries) All countries in the south-east of the WHO European Region including the Balkans, Turkey

and Israel

Data from Eurostat, Eurobarometer and Eurofound, which cover only the EU countries and a few additional countries from the European Free Trade Association or EU candidate countries, use the subregional distinctions of “EU15” (for the 15 Member States belonging to the EU before May 2004) and “NMS12” (for the 12 Member States joining the EU after May 2004). Total figures for all EU Member States are labelled “EU27”.

For all figures and tables in this report, subregional terms such as “EU15” or “Euro 2”, for example, indicate that all the respective countries in these subregions are covered by the data. If data from one or more countries are missing, subregional terms “EU15 countries” or “Euro 2 countries” are used instead, indicating that the data are not based on all the countries within the respective subregion.

5 Euro 1: EU countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands, Portugal, Spain, Sweden, United Kingdom� Non-EU countries: Andorra, Iceland, Monaco, Norway, San Marino, Switzerland� Euro 2: Bulgaria, Cyprus, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Romania, Slovakia, Slovenia� Euro 3: Armenia, Azerbaijan, Belarus, Georgia, Kazakhstan, Kyrgyzstan, Republic of Moldova, Russian Federation, Tajikistan, Turkmenistan, Ukraine, Uzbekistan� Euro 4: Albania, Bosnia and Herzegovina, Croatia, Israel, Montenegro, Serbia, the former Yugoslav Republic of Macedonia, Turkey�

Introduction10

map 1. European subregions used for the assessment

Euro 1

Euro 2

Euro 3

Euro 4

The aggregation of data for the European subregions demonstrates another important restriction encountered during the assessment. Depending on the data source and the stratification of data, it was very difficult – and sometimes impossible – to derive accurate results for the subregions that would incorporate the population weight of the countries within the respective region. For many indicators, this would require a calculation of population subgroup sizes (by age group, sex, income, and so on) for each country to be used as a weighting factor in calculating the subregional average. In many cases – especially when combining several determinants – this proved impossible due to a lack of adequate data. As a result, the findings presented for the subregions often represent the arithmetic average of the national rates of the countries covered by the respective subregion, not adjusted for the different national population sizes. In each figure, this restriction is clearly marked as the average of national rates for all reporting countries of the subregion. Subregional data that are representative (often provided by Eurostat databases) do not include this indication.

dATA ACCESSSources of data are listed in the reference section of each chapter. National data tables downloaded from these sources in spring 2011 can be requested by email from the WHO European Centre for Environment and Health. Please send your requests to [email protected], marked “National EH inequality data tables”.

rEFErENCES

Blas E, Sivasankara Kurup A, eds. (2010). Equity, social determinants and public health programmes. Geneva, World Health Organization (http://whqlibdoc.who.int/publications/2010/9789241563970_eng.pdf, accessed 13 April 2011).

Blas E, Sommerfeld J, Sivasankara Kurup A, eds. (2011). Social determinants approaches to public health: from concept to practice. Geneva, World Health Organization (http://whqlibdoc.who.int/publications/2011/9789241564137_eng.pdf, accessed 13 April 2011).

Braubach M et al. (2010). On the way to Parma: understanding and addressing the influence that social inequities have on environmental health. European Journal of Public Health 20(1):12–13.

CSDH (2008). Closing the gap in a generation: health equity through action on the social determinants of health. Final report of the Commission on Social Determinants of Health. Geneva, World Health Organization (http://whqlibdoc.who.int/publications/2008/9789241563703_eng.pdf, accessed 13 April 2011).


Dahlgren G, Whitehead M (2006). European strategies for tackling social inequities in health: levelling up, part 2. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0018/103824/E89384.pdf, accessed 30 November 2011).

Ministry of Social Affairs and Health, Finland (2006). Health in all policies: prospects and potentials. Helsinki, Ministry of Social Affairs and Health (http://ec.europa.eu/health/archive/ph_information/documents/health_in_all_policies.pdf, accessed 22 November 2011).

WHO (1946). Constitution of the World Health Organization. Geneva, World Health Organization (http://whqlibdoc.who.int/hist/official_records/constitution.pdf, accessed 28 November 2011).

WHO (1978). Declaration of Alma-Ata. Geneva, World Health Organization (http://www.who.int/hpr/NPH/docs/declaration_almaata.pdf, accessed 28 November 2011).

WHO (2009a). World Health Assembly Resolution WHA62.14. Reducing health inequities through action on the social determinants of health. Geneva, World Health Organization (http://apps.who.int/gb/ebwha/pdf_files/A62/A62_R14-en.pdf, accessed 28 November 2011).

WHO (2009b). Environment and health risks: the influence and effects of social inequalities. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0020/115364/E93037.pdf, accessed 13 April 2011).

WHO (2010a). Social and gender inequalities in environment and health. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0010/76519/Parma_EH_Conf_pb1.pdf, accessed 13 April 2011).

WHO (2010b). Environment and health risks: a review of the influence and effects of social inequalities. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0003/78069/E93670.pdf , accessed 13 April 2011).

WHO (2010c). Parma Declaration on Environment and Health. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0011/78608/E93618.pdf, accessed 13 April 2011).

WHO (2010d). Towards environmental health inequality reporting. Report of an expert group meeting, Bonn, Germany, 25–26 October 2010. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0013/130243/e94628.pdf, accessed 13 April 2011).

WHO (2011a). Rio Political Declaration on Social Determinants of Health. Geneva, World Health Organization (http://www.who.int/sdhconference/declaration/Rio_political_declaration.pdf, accessed 22 November 2011).

WHO (2011b). Closing the gap: policy into practice on social determinants of health: discussion paper. Geneva, World Health Organization (http://www.who.int/sdhconference/Discussion-Paper-EN.pdf, accessed 22 November 2011).

1

THE CONCEPT OF ENVIRONMENTAL

HEALTH INEQUALITIES

CHAPTER

Chapter 1. The concept of environmental health inequalities14

CHAPTEr 1. THE CONCEPT OF ENvIrONmENTAl HEAlTH INEquAlITIESGeorge Morris, Matthias Braubach

A HISTOrIC PErSPECTIvEWhile priorities and emphases change over time and according to location, the environmental health approach has traditionally centred on protection of population health through identifying, monitoring and controlling the environmental hazards which produce disease in populations. The approach has its origins in the earliest days of the modern public health movement and, by assuring the quality of domestic, community and occupational environments, has greatly extended lifespans and improved health and well-being for communities and individuals. Underpinned by advances in epidemiology and the biological understanding of disease, the disease-centred, hazard-focused approach to environmental health remains a cornerstone of public health activity.

It was clear even to the 19th-century public health pioneers that the degraded, malodorous neighbourhoods where lives were shortest and most blighted by disease were also home to the poorest communities. Recognition of the importance of environmental conditions for population health has always been, and continues to be, accompanied by recognition of the interplay between sociodemographic and physical factors in producing inequalities in health and well-being. The final report of the CSDH, Closing the gap in a generation (CSDH, 2008), reinforces the global relevance of this interplay for the 21st century. Notably, the first of the report’s three principles of action to tackle social inequity in health is: “improve the conditions of daily life – the circumstances in which people are born, grow, live, work, and age”. The report is suffused with references to the alignment of, and interplay between, sociodemographic, economic and physical factors in ways which bear on health and equity. This reinforces the fact that the physical environment – alongside the social environment and genetic endowment – is one key driver in the creation and destruction of health and well-being, and thus also a main driver for health inequalities.

In summary, in the 150 years and more during which there has been tangible interest in population health and action at the level of society to protect and improve it, inequalities in health between different social groups have been an abiding public health challenge. As the second decade of the 21st century begins, it could be argued that remarkable health gains delivered by adherence to a population focus are increasingly overshadowed by persisting and increasing variability between and within countries. Implicitly, the notable achievements of public health are not enjoyed by all. By extension, and despite a sometimes diminished political profile, differences in the physical context for people’s lives – referred to as environmental justice issues or environmental health inequalities – remain central to the health inequalities challenge throughout the world.

Acutely aware of this issue, ministers and representatives of Member States of the WHO European Region came together in 2010 at the Fifth Ministerial Conference on Environment and Health in Parma, Italy, to chart the next steps in the European environment and health process and, in the words of the Parma Declaration, “to face the key environment and health challenges of our time” (WHO, 2010).


Listed among the key environment and health challenges on which they made a commitment to act were: “the health risks to children and other vulnerable groups posed by poor environmental, working and living conditions (especially the lack of water and sanitation)” and “socioeconomic and gender inequalities in the human environment and health”. Each of these reflects recognition at a political level of the interplay between social and environmental variables and the need to tackle each to create better and more equal health for all. Evidence providers in the scientific, medical and epidemiological communities also recognize shared agendas and methodological challenges and the need to work in partnership. There is, for example, both conceptual and methodological overlap between the activities of environmental epidemiologists who study the effects of environmental exposures on health and disease in the population and social epidemiologists who often use social concepts to better understand and explain patterns of health in the population.

Key message 1Sociodemographic inequalities in the exposure to environmental hazards exist everywhere and they are not new. These inequalities can be expressed in relation to factors such as income, education, employment, age, gender, race/ethnicity and specific locations or settings.

INEquAlITy ANd INEquITyNotions of fairness versus unfairness and justice versus injustice now inform the language of public health when speaking about health differences between population subgroups. Like any type of inequality, inequalities in health and its determinants between different groups of people may, on one level, be regarded as natural and inevitable. An example might be differences in a range of health outcomes observed between different demographic groups, such as the elderly versus the rest of the population. However, normally, when health inequalities are observed between socially defined groups, they are more accurately described as “health inequities”. This term is now widely used to denote situations in which the distribution of health and its determinants is not simply unequal, but also unjust, unfair and avoidable (WHO, 2011).

Viewed from an environmental health perspective, the differential exposure of groups of people to health-relevant aspects of environment (with potential to create and sustain differences in health status) can often simply be inequalities. This might be the case where a group of people chooses to live in a polluted city centre for reasons of convenience or chooses riverside homes – potentially more liable to flooding – for aesthetic reasons or social status. However, the potential for differences in health outcome linked to environment may have little or nothing to do with choice or biological variation and may have its origin in factors beyond the influence of those affected. Here the environmental health challenge is about addressing health inequities that are unfair and avoidable.

Key message 2 The term “health inequalities” refers to general differences in health. Many of these differences (particularly where they are linked to social variables or gender) represent “health inequities” because they are unfair, unjust and avoidable.

IdENTIFICATION ANd ASSESSmENT OF INEquAlITIES ANd INEquITIESTo identify, assess, monitor and ultimately address environmental health inequalities and inequities associated with sociodemographic determinants, it is necessary to develop appropriate measures of environmental quality; this includes health-promoting aspects of environment in relation to social or demographic variables such as income, education, employment, age, gender, race/ethnicity and


specific locations or settings. In practice, this creates a requirement for robust indicators to illustrate the relationship between environmental health risk and different sociodemographic variables and permit better understanding regarding specific risk groups and their exposure. Such indicators might be termed “indicators of environmental health inequality”. Thus, the concentration of, for example, particulate matter in different areas might be related to the income of people living there to create an indicator of environmental health inequality. The potential to use such indicators to better understand public health problems and to shape and evaluate the policy response is considerable. They permit an assessment and analysis of the social distribution of exposure and impacts and allow comparison within and between nations which, when used constructively, can benefit all concerned. When used to assess environmental health inequality within nations, such inequality indicators can tease out problems which might otherwise be masked by average figures. Clearly, the importance a country attaches to delivering better life circumstances for its most disadvantaged groups, and its success in doing so, can be considered a telling indicator of its political, social and economic development. This reinforces the value of indicators of environmental health inequality in a broad context.

Nevertheless, indicators have no practical value unless they can be used to gather and process information about environmental health inequalities in practice, and this demands data. In many locations the data are simply absent, perhaps due to a lack of political will or resources to create the systems and structures for data generation. This may, of itself, be an inequality.

Key message 3Robust indicators of environmental health inequality that combine both social and environmental factors are needed, allowing these inequalities to be identified, assessed and tackled.

ENvIrONmENTAl HEAlTH INEquAlITy: HAzArd ANd rISKEnvironmental health inequalities can result in many ways: a typical mechanism is when areas populated by particular social groups have a greater concentration of environmental hazards and a scarcity or absence of environmental “goods”. This will almost inevitably disadvantage or marginalize population subgroups with certain characteristics in relation to, for example, gender, ethnic origin, occupation, income level, urban versus rural location, and so on. This unequal distribution between social groups is frequently described as an absence of “environmental justice” (Bullard, 2008; Curtice et al., 2005). However, the “distributive” element of environmental justice (which is about achieving more equal distribution of hazards and goods between population groups) is inextricably bound up with a need to ensure that different groups have equal capacity to influence decisions about what is and is not situated in their area. This second component of environmental justice is often termed “procedural justice” and a lack of procedural justice, in addition to the lack of distributive justice, often characterizes sociodemographically disadvantaged groups.

However, a comprehensive consideration of the role of sociodemographic factors in environmental health inequality must look above and beyond individual components of environmental justice to consider how a range of sociodemographic variables can modify not only the presence or absence of environmental hazards but also the individual risk of exposure and associated health consequences. Looking more closely at the influence of sociodemographic factors on environmental risk for the individual, it can readily be appreciated that factors such as age, gender, SES and indeed culture may have quite a profound influence on whether an individual chooses to be physically active or not, or to behave in a way which results in higher or lower exposure to a hazardous aspect of environment. There is a need to recognize that where sociodemographic factors influence individual opportunity, empowerment and dignity, they may critically influence an individual’s decision, for example, to be physically active or avoid harmful exposures. It is evident that there is a socially-mediated mechanism that affects the individual risk exposure. In any case, all dimensions of environmental health inequality are of equivalent policy relevance.


Key message 4 There are sociodemographically determined inequalities in population-level environmental hazard but, because sociodemographic factors may also modify individual exposure and the health impact for the same degree of exposure, there may also be sociodemographically determined inequalities in individual environmental health risk. Each level needs to be identified and assessed.

SOCIOdEmOGrAPHIC FACTOrS, ExPOSurE ANd vulNErAbIlITyAs described above, inequalities in environmental conditions and the lack of distributive justice regarding the location of environmental hazards lead to a greater probability of exposure to environmental health threats. Indeed, a wide range of surveys have shown that marginalized and disadvantaged groups – irrespective of the type of disadvantage, which can be education- or income-related as well as gender-specific or associated with ethnicity – are most often characterized as having the highest levels of exposure to environmental problems.

It is further recognized that the same degree of environmental exposure can result in a greater health impact when borne by a disadvantaged population; this may be due to a lower ability to respond to the environmental stress, perhaps exacerbated by other health pressures leading to synergistic effects. For example, sociodemographic and other factors can influence whether, having been exposed to a health-determining environmental factor, an individual goes on to experience a particular health outcome, and to what extent (the “exposure–response function”). The capacity for age, gender, genetic inheritance, pre-existing illness or psychosocial stress – singly or in joint interaction – to influence the exposure–response function is well understood in many cases.

Key message 5 Sociodemographic inequalities can be caused by differences in exposure to environmental risks (exposure differential), as well as by social or demographic differences in vulnerability towards certain risks (vulnerability differential).

THE PSyCHOSOCIAl dImENSION IN ENvIrONmENTAl HEAlTH INEquAlITyIn recent times, observation of and scientific interest in the vulnerability differential between different sociodemographic groups have led researchers to consider the role played by psychosocial stress. Psychosocial stress may have a number of origins but may be directly linked to social and physical characteristics of the places where people live. It is increasingly seen as a key factor in determining individual vulnerability to environmental hazards (Gee, Payne-Sturges, 2004). Psychosocial stress produces acute and chronic changes in the functioning of body systems such as those governing immune and inflammatory response, leading directly to illness or perhaps rendering individuals more vulnerable when exposed to, for example, a toxic environment. This may be particularly important in the absence of any counterbalancing effect from positive life circumstances and resources. Reflecting on the significant health inequalities challenge in Scotland and on the role of the physical environment, Scotland’s Chief Medical Officer emphasized the importance of the psychosocial dimension when he observed that “how people feel about their physical surroundings, can impact on not just mental health and well-being, but also physical disease” (Scottish Government, 2007).

Key message 6People’s perceptions of the physical aspects of the places they live in can profoundly impact on their mental and physical health and their longevity.


FrAmING THE PrOblEmDuring the meeting of an expert group on environmental health inequalities held in Bonn in September 2009 (WHO, 2009), attendees discussed a WHO contextual framework (see Fig. 2) which seeks to structure and identify potential pathways through which sociodemographic variables influence:• the nature and distribution of environmental conditions;• the exposure to these conditions for individuals and groups within society;• the exposure–response relationships which lead to different health outcomes in individuals with

comparable exposures;• access to, and quality of, health-related services.

The framework offers a useful, holistic approach to framing issues in environmental health, reflecting growing enthusiasm for more integrated policy-relevant strategies for analysing complex multifactorial environmental health issues. This tendency is stimulated by the increasing aspiration to anticipate and mitigate the likely health implications of policies in all sectors of civil society (sometimes referred to as the HiAP agenda). The framework attempts to make explicit to policy- and other decision-makers that better, more equal environmental health for individuals and communities requires attention to environmental factors and sociodemographic conditions, as well as to how these influences interrelate.

The framework suggests four major pathways through which sociodemographic inequalities may influence exposure to and health outcomes from environmental risks.• Arrow 1: there is a relationship between sociodemographic determinants and environmental

conditions. Disadvantaged groups may live and work in, or be surrounded by, less favourable environmental conditions than the general population, resulting in higher exposure risk.

• Arrow 2: factors attributed to sociodemographic inequalities (such as knowledge and health behaviour) compound exposure. Given the same environmental conditions, disadvantaged groups may be more exposed than the general population.

• Arrow 3: factors attributed to sociodemographic inequalities (such as health status and biological sensitivity) influence the exposure–response function. Given the same exposure, disadvantaged groups may be more vulnerable to adverse health effects than the general population.

• Arrow 4: sociodemographic inequalities have a direct impact on health outcomes, which may operate through many mechanisms – some environmental, some independent of environmental factors. However, given the same exposure–response situation, disadvantaged groups may also be more vulnerable to adverse health effects than the general population (through, for example, inadequate insurance, reduced health services use or reduced access to services).

Arrows 1 and 2 together represent the exposure differential, describing the increased exposure risk, while Arrow 3 represents the vulnerability differential, accounting for an increased translation of environmental exposure conditions into negative health effects.

The generic framework may be populated for different environmental health issues but the utility of Fig. 2 is perhaps best illustrated by example. An ostensibly straightforward environmental health issue, such as the risk of serious health effects and even death from carbon monoxide (CO) poisoning in the home, can be expanded using the framework to illustrate its many components and the array of policy options it presents.

Briefly, particular sociodemographic groups – such as the elderly and those on a low income – may be more likely to occupy low-quality housing, possibly with poorly maintained gas or solid fuel heating systems, thus illustrating the potential for sociodemographic factors to create differential exposure to CO. The elderly and cognitively impaired and those with limited education may, in turn, possess less knowledge or understanding of the hazard of CO, yet may actually spend longer in the home environment and, in a struggle to keep warm on a limited income, may deliberately limit ventilation to save on fuel costs. These additional factors illustrate the potential influence of sociodemographic


variables on exposure to CO, and hence the risk to an individual or demographic group. To further compound the risk, the elderly may exhibit particular vulnerability to the health impacts of CO due to pre-existing illness or impaired cardio-respiratory function, thus illustrating that sociodemographic factors may also influence individual vulnerability. Finally, sociodemographic factors may align with factors such as inadequate health insurance, or reduced access to or use of health services, which may affect the effectiveness of medical treatment.

Fig. 2. The wHO framework model on social inequalities and environmental risks

Exposure Health effects and costs

Exposure–response function

Stakeholders and HiAP actors (environment, housing, transport, social, etc.)

Individual susceptibility

Access to/quality of health services

Public, health and social services/ health system

Environmental protection

Preventive environmental and health services

Health protection/education

Effect modifiers

Environmental conditions

Driving forces Macroeconomic context: increasing social disparities and stratification

1 2 3 4

Inequalities by sociodemographic determinants (income, education, age, occupation, migrant status, gender, etc.)

Source: modified, based on WHO (2009).

Key message 7 It is important to take a holistic approach to framing equity problems in environmental and human health if the complexity of the challenge and the policy options are to be revealed.

CONCluSIONThere is strong justification for work to establish indicators and reporting systems that relate health-relevant environmental factors to individual or multiple sociodemographic variables with an appropriate resolution, such as at a small geographical scale. These indicators are a necessary resource for work to improve understanding, reporting, assessing and ultimately acting on environmental health inequalities and the wider challenge of social complexity in environmental public health. There is also strong justification for keeping under review, and extending where indicated, the range of environmental variables which should be linked to sociodemographic variables. Equity-sensitive environmental monitoring and reporting should in future extend its scope to include environmental factors which are potentially health-nurturing and represent a health resource rather than a health risk. Similarly, integration of aspects of physical environments or places which generate negative psychosocial responses is also important.


Increasing reference to the concept of “place” and “healthy places” in public health circles represents recognition of the interplay of social, demographic, economic and cultural factors at the neighbourhood level and the need to ensure that these are jointly managed to create better and more equal health. Such an approach is a significant enrichment of the hazard-focused traditions of environmental public health.

Finally, the current environmental changes and erosion of natural resources linked to climate change, and the policies put in place to mitigate and adapt to them, are likely to impact differentially on sociodemographic groups, exacerbating environmental health inequalities and further reinforcing the case for appropriate indicators and reporting structures of environmental health inequality.

The development and application of environmental health inequality indicators for the WHO European Region – as presented in this report – permits a first baseline assessment of environmental health inequalities. This, in turn, has the potential to inform policy and action and to chart progress in tackling the environmental dimension of sociodemographic inequalities in health.

Key message 8The range of environmental health inequality indicators should continue to reflect the nature of environmental health threats. The assessment of environmental inequalities, therefore, should be dynamic and based on frequently updated inequality indicators.

rEFErENCES

Bullard RD (2008). Environmental justice in the 21st century. Atlanta, GA, Environmental Justice Resource Center at Clark Atlanta University (http://www.ejrc.cau.edu/ejinthe21century.htm, accessed 3 October 2011).

CSDH (2008). Closing the gap in a generation: health equity through action on the social determinants of health. Final report of the Commission on Social Determinants of Health. Geneva, World Health Organization (http://whqlibdoc.who.int/publications/2008/9789241563703_eng.pdf, accessed 5 June 2011).

Curtice J et al. (2005). Public attitudes and environmental justice in Scotland: a report for the Scottish Executive on research to inform the development and evaluation of environmental justice policy. Edinburgh, Scottish Executive (http://www.scotland.gov.uk/Resource/Doc/77843/0018790.pdf, accessed 5 June 2011).

Gee GC, Payne-Sturges DC (2004). Environmental health disparities: a framework integrating psychosocial and environmental concepts. Environmental Health Perspectives, 112(17):1645–1653 (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1253653/, accessed 18 September 2011).

Scottish Government (2007). Health in Scotland 2006: annual report of the Chief Medical Officer. Edinburgh, Scottish Government (http://www.scotland.gov.uk/Resource/Doc/203463/0054202.pdf, accessed 3 October 2011).

WHO (2009). Socioeconomic inequalities – scenarios, recommendations and tools for action. Background document for the Third High-Level Preparatory Meeting. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/Document/EEHC/29th_EEHC__Bonn_edoc15.pdf, accessed 15 October 2009).

WHO (2010). Parma Declaration on Environment and Health. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0011/78608/E93618.pdf, accessed 3 August 2011).

WHO (2011). Health impact assessment: glossary of terms used. Geneva, World Health Organization (http://www.who.int/hia/about/glos/en/index1.html, accessed 3 October 2011).

2

HOUSING-RELATED INEQUALITIES

CHAPTER

Chapter 2. Housing-related inequalities22

CHAPTEr 2. HOuSING-rElATEd INEquAlITIESSéverine Deguen, Lucia Isabel Fiestas and Denis Zmirou-Navier

Key messages Generally:• within the EU, the NMS12 countries are more affected by substandard housing than the EU15

countries;• for countries beyond the EU, no adequate data (except on water supply) were available, but

published literature indicates that the housing situation is worse than in the NMS12 countries;• inequalities have been found in all European countries with available data; • subgroups of low-income population and single-parent households have been identified as more

exposed to poor housing conditions in all countries;• poverty increases the level of inequality among single-parent households.

More specifically:• inadequate water supply is mainly an issue in Euro 3 and Euro 4 countries;• rural populations are more exposed to inadequate water supply than urban populations;• the lack of a flush toilet and bath or shower is most prevalent in less wealthy countries;• lower-income households, and particularly low-income single-parent households, are most likely

to lack a flush toilet and lack a bath or shower;• a low income increases the risk of living in overcrowded housing, particularly among single-

parent households;• living in a damp dwelling is a housing issue in both the EU15 and NMS12 countries;• low-income subgroups are more exposed to damp dwellings, especially in the NMS12 countries,

where a substantial gap exists between the lowest income quintile and the higher ones;• there is a high prevalence of inability to keep the home warm in most European countries,

especially among single-parent households and low-income population groups;• inability to keep the home cool in summer – even more prevalent in the population than inability

to keep the home warm in winter – also shows income-related inequalities, although they are less strongly expressed than those for keeping the home warm.


INTrOduCTIONAwareness of socioeconomic health inequalities has recently put this topic on the priority list when drafting public health policies. It is well known that populations privileged by advanced education, greater income, higher-status jobs and better quality housing standards are in better health. Housing and neighbourhood quality improvement has historically been a key policy instrument to improve health (Howden-Chapman, 2002). Housing in particular has been recognized as an important parameter affecting population well-being and health (Bonnefoy, 2007; Braubach and Fairburn, 2010; Poortinga, Dunstan and Fone, 2008). One reason for this is that a high proportion – approximately two thirds – of time is spent in the home (Bernstein et al., 2008). This proportion varies across countries and is even greater for vulnerable population subgroups such as the elderly, children – at 80–90% of the day (Breysse et al., 2004) – and deprived people, who may be unemployed or have few external activities. Furthermore, less affluent populations are more often affected by inadequate housing conditions (Braubach and Fairburn, 2010).

Inadequate housing conditions generate health inequalities through two main pathways. First, from an environmental epidemiology perspective, the term “housing conditions” may be used as a proxy for environmental exposures influencing health in the home, such as indoor air quality, noise or humidity. Second, it can be an indicator of socioeconomic determinants of health: poor housing conditions are a consequence of a disadvantaged situation. Housing prices vary greatly geographically and consequently the quality of housing and of its local environment is both directly and indirectly associated with income, or more generally with SES (Braubach and Fairburn, 2010; Galobardes et al., 2006).

It is important to note that exposure inequalities may occur at multiple levels: at an individual level (such as personal or household income, or substandard housing) and at a geographical, or ecological, level (such as average neighbourhood income, or proportion of inadequate housing in the neighbourhood). The analysis in this chapter focuses only on information regarding individuals and households.

Associations between inadequate housing and health have been revealed using a variety of approaches and indicators. Living conditions affect both physical and mental health. Data collected by WHO on eight European countries confirm that inadequate housing conditions are associated with risk factors such as moulds or overcrowding, especially in low-income households (Braubach and Fairburn, 2010). Several researchers have found associations between indoor environmental exposures including dampness and a variety of adverse respiratory diseases, such as risk of asthma, pulmonary infections and allergies (Krieger and Higgins, 2002; Sandel and Wright, 2006). Damp houses provide a nurturing environment for mites, roaches, respiratory viruses and moulds known to play an important role in respiratory disease pathogenesis. For this reason, poor indoor air has been advanced as a way of explaining the recent increase in prevalence of respiratory health problems in many developed countries (Rauh, Landrigan and Claudio, 2008).

In addition, overcrowding and lack of hygiene and sanitation equipment are relevant health hazards found in dwellings. Overcrowding has long been associated with the transmission of tuberculosis and respiratory infections. In recent years, epidemiological studies have revealed an increased risk of chronic diseases for those living in substandard housing (Krieger and Higgins, 2002) and documented that poor living environments – including overcrowding, inadequate garbage removal or location near busy transportation routes – may generate chronic stress (Rauh, Landrigan and Claudio, 2008).

Low indoor temperature is associated with lower health status, and particularly with cardiovascular disease (Krieger and Higgins, 2002). A synthesis of intervention studies reviewed several reports assessing changes in health following housing improvements (Thomson, 2011). These confirmed that housing intervention programmes targeting thermal comfort improved health – including mental health – and social outcomes considerably (Shortt and Rugkåsa, 2007). The association between degraded housing characterized by poor thermal efficiency and high levels of excess winter or summer mortality has also been documented (Braubach and Fairburn, 2010). Homes of low-income individuals are more likely to be too cold (or too warm in summer) because of insufficient insulation and lack of air


conditioning. In summary, substandard housing is not evenly distributed across space and population; disadvantaged groups are disproportionately affected. In some cases, low-income households even have to make tradeoffs between having enough food and living in adequate housing conditions (Krieger and Higgins, 2002).

dATA ANd mETHOdSData from the WHO/UNICEF Joint Monitoring Programme ( JMP) for Water Supply and Sanitation (2011) were used to analyse inequalities in inadequate water supply. Data are available for all 53 Member States of the WHO European Region, collected in five different years: 1990, 1995, 2000, 2005 and 2008. Population with inadequate water supply is defined as the percentage of the population supplied with water from unimproved sources (water supply that, by nature of its construction or through active intervention, is not protected from outside contamination – in particular from contamination with faecal matter – such as surface water or water from unprotected wells or springs). Unfortunately, the WHO/UNICEF JMP data available up to 2008 do not include data on individuals (such as gender or income) and only distinguish between urban and rural areas. However, as the recent JMP data collection wave (2010 data to be released soon) includes information on water and sanitation supply by income decile, more detailed assessment of these data will soon be possible.

A bar chart showing the proportion of the population with inadequate water supply for urban and rural areas within each country and subregion was created for the year 2008. Additionally, a time trend graph was drawn to represent the evolution of the prevalence of inadequate water supply in rural areas in the WHO European Region, to detect whether the situation had improved during the last 20 years.

For the water inequality indicator, the data were disaggregated by subregions Euro 1 to Euro 4 to present results for the different subregions.

Data for analysis of the five other housing inequality indicators were retrieved from EU-SILC database (2011). Data were available from 2004 to 2009 and were downloaded from March to May 2011. Information was reported by 30 countries in 2009, but the number of countries repeatedly reporting over time between 2004 and 2009 was lower, so it was not possible to assess time trends for a variety of countries. As a result, only data from the latest year, 2009, were studied. The five housing inequality indicators covered in this report are:• lack of flush toilet in the dwelling, based on EU-SILC question “Is there an indoor flushing toilet

in your dwelling?” with answer options Yes and No;• lack of bath or shower, based on EU-SILC question “Is there a shower unit or a bathtub in your

dwelling?” with answer options Yes and No;• overcrowding, based on the following Eurostat definition: “a person is considered as living in an

overcrowded household if the household does not have at its disposal a minimum of rooms equal to one room for the household, one room per couple in the household, one room for each single person aged 18 or more, one room per pair of single people of the same gender between 12 and 17 years of age, one room for each single person between 12 and 17 years of age and not included in the previous category, and one room per pair of children under 12 years of age”;

• dampness in the home, based on EU-SILC question “Do you have any of the following problems with your dwelling/accommodation: a leaking roof, damp walls/floors/foundation, or rot in window frames or floor?” with answer options Yes and No;

• inability to keep the home adequately warm, based on EU-SILC question “Can your household afford to keep its home adequately warm?” with answer options Yes and No.

To complement the inequality indicator on thermal comfort in winter, data from an EU-SILC rotation module question “Is the dwelling comfortably cool during summer time?” with answer options Yes and No (only asked in 2007) were also used and are discussed in an additional section.


The EU-SILC data were stratified by four socioeconomic characteristics: age, income level reflecting relative poverty status, household type and sex. The age variable was divided into three groups: less than 18 years, between 18 and 64 years and over 65 years. The income variable was divided into population below and above relative poverty status (defined as household income below or above 60% of the median income for the respective country). Upon specific request, Eurostat provided additional data on income quintiles for some of the housing inequality indicators. The household type variable had 16 categories, but only three were used in this analysis: “households with dependent children”, “single-parent households with dependent children” and “single households with one adult older than 65 years”. Analysis by sex stratification was not relevant because most households – except single and single-parent households – are composed of a somewhat similar number of males and females so that statistical analysis rarely results in significant differences.

For the five housing-related inequality indicators based on Eurostat data, two different metrics were calculated:• a crude metric, denoting the prevalence of a given problem per country or subregion and per

socioeconomic or demographic category – results are expressed as a percentage of households or population;

• a relative metric, denoting the ratio dividing the prevalence among one socioeconomic category by the prevalence among another – the higher the ratio, the greater the level of disparity between the two categories.

Given that ratio figures can be misleading, both crude and relative metrics were included in graphs in order to highlight the countries for which housing conditions constitute a major problem – those countries with both a high proportion of the population affected and large inequalities between specific subpopulation groups. Conversely, several countries provide challenging cases when an overall low prevalence of a given problem (such as lack of a bath/shower) affects the calculation of ratios, leading to high ratios when specific population subgroups are compared with the average population. Thus, evaluation must always consider the absolute and the relative dimensions of inequality.

Scatter plots were also used to illustrate the relationship of prevalence of one housing problem between two different population subgroups. This was done to explore whether inequalities in the first population group deviated from the second one, indicating inequalities in exposure. The y-axis was expressed on a logarithmic scale to improve the readability of the graph because several countries had prevalence levels close to 0%. A linear model was fitted to investigate how the two prevalence values were linearly associated as well as to quantify the strength of the association.

In addition, to explore whether prevalence of inability to keep the home warm by country deviates from the general pattern of national inequality, information was extracted from the EU-SILC database on the 2009 Gini index by country. This is a standard economic measure of income inequality that varies between 0 and 100: a value close to 0 signifies a low level of income inequality and a high value indicates a high level of income inequality in the country. The scatter plot illustrates the shape of the relationship between the two variables (Gini index and prevalence of inability to keep home warm).

As the housing inequality data provided by Eurostat covers the EU countries and Iceland, Norway and Switzerland only, EU15 and NMS12 were chosen as subregions. Iceland, Norway and Switzerland are presented separately.

rESTrICTIONS ANd dATA lImITATIONSThe main limitation of the available data on inequalities in adequate water supply is the absence of information on socioeconomic and demographic characteristics, precluding a detailed analysis of inequalities. The only stratification possible is by rural or urban population, although data for 2010 will shortly be published, enabling data analysis by income decile. This information was used as a


proxy for SES because there is consistent evidence that, on average, rural populations experience higher socioeconomic disadvantages than urban ones. A second limitation is that the absence of annual data did not allow for a more precise trend analysis. Having acknowledged these limitations, the main advantage of the database lies in the great number of countries for which this information was available, permitting a large geographical comparison.

Data on all other housing inequality indicators were provided by Eurostat, covering a maximum of 30 Member States. A study of the 53 countries of the WHO European Region would allow a larger spatial view of housing inequalities, possibly improving identification of the most exposed population groups in a more economically and culturally diverse area. Furthermore, the sample sizes for the EU-SILC survey are not very large, varying between 8250 (Germany) and 2250 (Iceland) households, which may result in unstable figures and large uncertainty levels when split according to several subcategories.6 Nevertheless, a detailed data search (WHO, 2011a) has indicated that these are the best available data to date, although caution should be used when interpreting the results of the analysis.

In particular, information on the composition of households, such as the number of people per age category, would be relevant to improve understanding of the overcrowding phenomenon in certain European subregions. This information might help to distinguish the countries for which overcrowding is related to cultural habits (such as cohabitation of several generations) from those for which it reflects a situation of deprivation.

Regarding analysis of dampness in the dwelling, some qualification of the degree of dampness would also be appropriate. The information used for the definition was rather poor and misclassification of exposure could result in over- or underestimation of the proportion of subjects exposed to damp dwellings, possibly in a different manner across countries.

Climate information is important in order to interpret fully analysis of the inequality indicator on thermal comfort – cold countries could be more affected by inability to keep homes warm. Perception of cold in dwellings might also differ from one group to another; for example, older subjects or unemployed people who spend more of the day inside their home could be more vulnerable to perception of cold compared to those who are out for a greater part of the day. Data on indoor temperature could allow stratification of the analysis and enhance identification of the most exposed populations. In addition, information on the number of rooms per dwelling could also help to explain differences in the prevalence of inability to keep the home warm between countries and subgroups of population.

For all the housing inequality indicators, it would also be of great interest if additional socioeconomic information (such as education level and employment status) were routinely collected as a way to improve the understanding of inequality factors between and within countries.

INEquAlITIES IN INAdEquATE wATEr SuPPly

IntroductionAn adequate supply of water is recognized as a basic human need as well as a human right (United Nations Human Rights Council, 2010). Therefore, halving the proportion of individuals without access to safe drinking-water is one of the key Millennium Development Goals (United Nations Millennium Project, 2005). The need for water resources goes beyond quantity and must also consider quality for maintaining good health. The United Nations Development Programme (UNDP) declared in March 2011 that:

6 For details on EU-SILC sample size and methodology see Eurostat website (http://epp�eurostat�ec�europa�eu/portal/page/portal/income_social_inclusion_living_conditions/introduction)�


Water and sanitation infrastructure in many countries in Europe and Central Asia are in a critical state and deteriorating, often posing a threat to human health. Yet, most countries have signed or ratified core UN human rights conventions, and many regional treaties recognize the right to water…. The internationally recognized right to water guarantees that all people have access to safe potable water at an affordable price, regardless of their age, sex, race, gender, or ethnicity.

(UNDP, 2011)

Drinking-water has been defined by WHO as water that does not contain pathogens or chemical agents at levels of concentration that could affect health (WHO 2011b). The greatest hazard associated with drinking-water is contamination by sewage or by human excrement. It is also significant that the UNDP declaration goes on: “The right to water emphasizes the importance of water-related development for marginalized and vulnerable groups, who are commonly socially excluded”. In this context, a forthcoming report by the United Nations Economic Commission for Europe (UNECE), WHO and the French Ministry of Health on good practices to ensure equitable access to water and sanitation in the pan-European region concludes that the three major dimensions included in the concept of equitable access to water and sanitation are geographical disparities in service provision, discrimination or exclusion in access to services by vulnerable and marginalized groups, and financial affordability by users (UNECE, in press).

Fig. 3. Prevalence of inadequate water supply by urbanization level (2008) in countries without full coverage

0

5

10

15

20

25

30

35

40

45

Por

tuga

lG

reec

eE

uro

1 [a

]

Slov

enia

Est

onia

Latv

iaE

uro

2 [a

]

Turk

men

ista

nB

elar

usU

krai

neG

eorg

iaA

rmen

iaK

azak

hsta

nR

ussi

an F

eder

atio

nK

yrgy

zsta

nR

epub

lic o

f Mol

dova

Uzb

ekis

tan

Aze

rbai

jan

Tajik

ista

nE

uro

3 [a

]

MK

D [b

]Al

bani

aB

osni

a an

d H

erze

govi

naS

erbi

aC

roat

iaM

onte

negr

oTu

rkey

Eur

o 4

[a]

All

coun

tries

[a]

Pre

vale

nce

(%)

Urban population with inadequate water supplyRural population with inadequate water supply

Source: data from WHO/UNICEF JMP, 2011.Notes: [a] data for Euro 1, 2, 3 and 4 represent the average of national rates for all countries within the subregions, including those with full coverage not shown; [b] MKD: International Organization for Standardization (ISO) code for the former Yugoslav Republic of Macedonia.

Indicator analysis: inequalities by urbanization levelThe proportions of rural and urban populations with inadequate water supply, defined as “unimproved water supply conditions” within the WHO/UNICEF JMP database, were compared (see Fig. 3). The distribution of the prevalence of inadequate water supply in rural and urban sectors varies across countries from 0% in most Euro 1 and Euro 2 countries to about 40% for the rural population living in Tajikistan, a Euro 3 country. Euro 3 countries are by far the most affected by inadequate water supply, followed by Euro 4 countries. However, water supply inequalities were also found among some Euro 1


and Euro 2 countries (Euro 1: Portugal and Greece, Euro 2: Estonia and Latvia). Overall, Azerbaijan has the highest proportion (12%) of urban population with inadequate water supply, whereas Tajikistan has the highest proportion (39%) of rural population without adequate access to water. However, several countries within the Euro 3 and Euro 4 regions present a low prevalence of inadequate water supply in rural and/or urban populations, with figures comparable to Euro 1 and Euro 2 countries. For instance, in Israel the entire population – urban and rural – has adequate access to water, while in Belarus and the former Yugoslav Republic of Macedonia only 1% of the rural population lacks adequate access to water. In terms of rural–urban inequalities, the largest differences are observed for Kyrgyzstan, Kazakhstan and Uzbekistan. For example, in Kyrgyzstan the proportion of the population without an adequate water supply is 15 times higher among the rural population than the urban population.

Inequality trends in inadequate water supply in urban and rural areas between the four Euro regions from 1990 to 2008 were also analysed (see Fig. 4). Already low, the prevalence of inadequate water supply in Euro 1 and Euro 2 countries slowly decreased between 1990 and 2008 from 0.7% to 0% (Euro 1) and from less than 2% to less than 1% (Euro 2) among the rural population. A similar pattern is observed among the urban population. In Euro 3 and Euro 4 countries, the proportion of the rural population with inadequate water supply increased between 1990 and 1995 and then decreased from 2000. However, the time trend variations are lower in the Euro 4 countries than the Euro 3 countries. In the Euro 3 region, the prevalence in 2008 is still higher for both urban and rural areas than the 1990 figure.

Fig. 4. Trends of inadequate water supply by urbanization level

Urban

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

1990 1995 2000 2005 2008Year

Pre

vale

nce

(%)

Euro 1 Euro 2

Rural

0

3

6

9

12

15

18

21

24

1990 1995 2000 2005 2008Year

Pre

vale

nce

(%)

Euro 3 Euro 4 All countries

Source: data from WHO/UNICEF JMP, 2011.

A more in-depth analysis (data not shown) by country highlights that the biggest improvement in water supply equality has been observed in Turkmenistan, where all the rural population had access to adequate water in 2008 while 28% had an inadequate water supply in 1995. Large urban–rural equality improvements were also found in Georgia, Tajikistan and Azerbaijan. A worsening trend between 1995 and 2008 was noted for Kazakhstan (2% increase in inadequate water supply in rural areas from 1995 to 2008) and in the Republic of Moldova and in Uzbekistan (4% increase in both cases).

Target groups for actionThese data show a clear difference across the WHO European Region, particularly between the Euro 1 and Euro 2 countries and the Euro 3 and Euro 4 countries. Many countries could qualify as “exposed” to the problem of inadequate water supply given the high prevalence of unimproved water sources in the general population: the situation in Azerbaijan and Tajikistan illustrates this point. Moreover, it was seen that in every country (except in the Euro 1 region, in which inadequate water supply is not an issue)


rural populations constitute a disproportionately highly exposed group compared to urban populations. Finally, for the Euro 1, Euro 2 and Euro 4 regions, the prevalence tended to decrease between 1990 and 2008 in both rural and urban areas. However, the prevalence remains very high for Euro 3 countries in 2008: higher than the figure for 1990.

National data sets providing more detailed information need to be applied to identify further inequalities. One example is contributed by Hungary, showing the provision of piped water supply to dwellings in relation to the rate of Roma population in Hungarian municipalities (see Annex 1).

Health implicationsA key driver for the provision of safe water is improvement to public health, and many studies have demonstrated that drinking-water interventions can substantially reduce the risk of diarrhoeal diseases (Clasen et al., 2007; Fewtrell et al., 2005). Diarrhoeal disease is one of the leading causes of morbidity and mortality in less developed countries and in less developed areas of emerging countries, especially among children. Rural areas are particularly at risk. Inadequate water supply translates into a major health impact as it leads to a high prevalence of water-borne diseases. In the WHO European Region, an average of 330 000 cases of serious water-related diseases are reported every year, including campylobacteriosis, viral hepatitis A, giardiasis, Shigella (bloody diarrhoea), enterohaemorrhagic Escherichia coli infection, legionellosis and cholera (WHO, 2011b). Increased knowledge about the preventable health burden associated with an inadequate supply of drinking-water has strengthened the commitment to increase access to improved water supplies across Europe. Improvements in access to safe water and adequate sanitation could reduce the child mortality rate by 2.2 million each year.

Rural area dwellers are particularly exposed to inadequate drinking-water sources; they face increased vulnerability to water-borne diseases due to lack of access to a water distribution network or to distance from water sources. Moreover, water service costs are comparatively higher in rural and suburban areas than in urban centres. This is due in part to the numerous intermediaries in the retail chain, and in part to the scarcity of the water. This situation translates into a major health impact for rural populations.

Conclusions and suggestionsThe two major conclusions of the analysis are that inadequate water supply is mainly an issue in Euro 3 and Euro 4 countries and that populations living in rural areas, particularly in these regions, are more exposed to an inadequate water supply. Improved water supply can enhance health status by enabling better hygiene, and possibly also by decreasing the need for storage in the home and for transport of water – factors that are linked to the risk of water contamination. In addition to quantity, it is well established that quality of drinking-water is an important determinant in preventing diarrhoeal diseases.

Suggested mitigation actions are:• protection of ground water or effective treatment and disinfection of surface water, and building

of water networks that bring protected drinking-water to remote areas;• linking provision of drinking-water with sanitation improvements to protect water resources

and reduce the risk of contamination;• in rural areas, investment by local and state authorities in community utilities for production

and distribution of improved drinking-water and provisions to ensure its permanent microbial and chemical quality, since it has been shown that poor water system maintenance, even for a few days, can jeopardize the expected health benefits (Hunter, Zmirou-Navier and Hartemann, 2009);

• better auditing by funders of water quality programmes of whether interventions are sustainable and whether health benefits are being achieved – this calls for health surveillance efforts to monitor the evolution of diarrhoeal diseases over time.


INEquAlITIES IN lACK OF A FluSH TOIlET IN THE dwEllING

IntroductionAccess to safe sanitation facilities and practices is central to human health and dignity, economic well-being, educational opportunities, and sustainability of the environment (Water Supply and Sanitation Collaborative Council, 2010). Inadequate access to safe sanitation services, together with poor hygiene practices, produces a compounded detrimental effect on the health of individuals and may lead to impoverishment via a severe decrease in economic and educational opportunities (UNICEF, 2011). Lack of access to a flush toilet in the dwelling may force individuals to defecate in the open or to use unsanitary facilities (Humphrey, 2009). Open defecation has been described by WHO as “the riskiest sanitation practice of all” (WHO, 2010a) because it constitutes a major cause of ground water pollution, agricultural produce contamination and disease transmission (Water Aid, 2010). Other excreta disposal options in use, including traditional pit latrines, “flying toilets”, ventilated pit latrines and pour-flush toilets, are also considered to be unimproved sanitation practices because they are shared by many households and pollute the groundwater through “direct and indirect discharge of pollution loads into the environment” (Katukiza et al., 2010). Poverty is still one of the major distal determinants of diarrhoea, and is closely related to lack of sanitation, poor neighbourhood infrastructure and poor living conditions (Genser et al., 2008): 40% of the world’s population – the most socioeconomically disadvantaged and materially deprived – do not have access to improved sanitation facilities (WHO/UNICEF JMP, 2010). Furthermore, the disparities between rural and urban communities are alarming. Without a flush toilet in the dwelling, the most impoverished populations are systematically prevented from experiencing the protective health effects of improved sanitation.

Indicator analysis: inequalities by income and household typeLack of a flush toilet in the dwelling is not a major issue for the EU15 countries (see Fig. 5). The average percentage of the population lacking a flush toilet in the dwelling is 0.7% for these countries, with Portugal reaching the highest prevalence at 2.4%. Every household in Denmark, Netherland, Spain, and Sweden reported the presence of a flush toilet in the dwelling. Similar results were also observed for Iceland, Norway and Switzerland. By contrast, wide variability was observed between the NMS12 countries. Many countries exhibit a very high percentage of the population lacking a flush toilet in the home – for instance, Romania (42.4%), Bulgaria (26.2%), Lithuania (17.2%) and Latvia (16.6%) – whereas the percentage is very low in other countries, including Slovenia (0.6%), Cyprus (0.7%) and the Czech Republic (0.7%). The entire population of Malta was reported to have a flush toilet in the dwelling.

Fig. 5 also shows the level of inequality within countries: the large majority of countries exhibit an income ratio greater than 1 when comparing the prevalence of households below the poverty threshold with those above it. This demonstrates that the prevalence of lacking a flush toilet is higher among populations living in relative poverty. Higher ratios are reported in the NMS12 than in the EU15 countries: about 17:1 and 15:1 for Slovenia and Cyprus, respectively, versus about 4:1 in Belgium, Finland, Greece and Luxembourg.

In conclusion, analysis of Fig. 5 highlights three main issues. First, it demonstrates that lack of a flush toilet is a significant housing problem in several of the NMS12 countries. Second, it exposes large inequalities between the socioeconomically privileged and disadvantaged populations in some countries where the baseline prevalence is low, such as Slovenia and Cyprus. Third, it highlights the compounded problem that exists in countries that present a high level of inequality in conjunction with a high prevalence of lack of a flush toilet in the home, as in Latvia and Hungary. Although the data presented here are restricted to EU members and Iceland, Norway and Switzerland, similar – or even stronger – income-related inequalities appear outside the EU, as shown in data on the prevalence of lack of a flush toilet stratified by wealth status, urban/rural residence and region in Georgia (see Annex 1).


Fig. 5. Prevalence of lack of a flush toilet in the dwelling by relative poverty level (2009)

0

10

20

30

40

50

60

70

80

90

Den

mar

k [a

]N

ethe

rland

s [a

]S

pain

[a]

Sw

eden

[a]

Italy

Irela

ndU

nite

d K

ingd

omB

elgi

umFr

ance

Finl

and

Luxe

mbo

urg

Ger

man

yA

ustri

aG

reec

eP

ortu

gal

EU

15

Mal

ta [a

]C

zech

Rep

ublic

Cyp

rus

Slo

veni

aS

lova

kia

Pol

and

Hun

gary

Est

onia

Latv

iaLi

thua

nia

Bul

garia

Rom

ania

NM

S12

EU

27

Nor

way

[b]

Sw

itzer

land

[b]

Icel

and

Pre

vale

nce

(%)

0

2

4

6

8

10

12

14

16

18

Rat

io o

f pre

vale

nce

(bel

ow:a

bove

re

lativ

e po

verty

leve

l)

Above relative poverty levelBelow relative poverty levelIncome ratio

Source: data from EU-SILC, 2011.Notes: [a] countries reporting full population coverage; [b] countries reporting 0.1% for both above and below relative poverty level.

The relationship between lack of a flush toilet in the dwelling among single-parent households and among the general population was plotted on a logarithmic scale (see Fig. 6). The analysis excludes countries where the entire population has a flush toilet inside the home. The linear model explains about 96% of the variability, signifying that one variable could predict the second one with very good precision. The regression coefficient is equal to 0.93, showing that, on average, the percentage of single-parent households living in a home without a flush toilet is slightly lower than that in the general population. Several exceptions appear on the scatter plot: countries on the left of the line have a higher prevalence of lack of a flush toilet among single-parent households than in the general population. The highest levels of inequality faced by single-parent households exist in Poland (4.8% of the general population reported lack of a flush toilet at home compared to 7.9% of single-parent households), Finland (0.8% versus 1.9%), Slovenia (0.6% versus 1.6%) and Lithuania (17.2% versus 20.3%).

The subgroup of single-parent households lacking a flush toilet in the dwelling was explored in more detail to assess whether those below the poverty threshold constituted a particularly exposed group. Using a linear model (data not shown), it was found that, on average, the percentage of single-parent households below the poverty threshold without a flush toilet increases by about 2% when the prevalence among all single-parent households increases by 1%. In other words, the prevalence of lack of a flush toilet in the dwelling is twice as high among single-parent households in relative poverty than among all single-parent households. The situation represented by several NMS12 countries appears especially worrying. For example, in Bulgaria the prevalence of living in a dwelling without a flush toilet is 19.2% among all single-parent households and 53.6% among those in relative poverty. In Romania, the prevalence is 42% among all single-parent households but almost double that figure (78%) among those in relative poverty. The examples for Romania and Bulgaria show how the combination of two sociodemographic determinants – namely poverty and single-parent household situation – enables the identification of a particularly exposed group.

Finally, the income-related inequality of lacking a flush toilet in the dwelling was analysed (see Fig. 7), indicating that low income is associated with increased prevalence of lack of a flush toilet. As already shown, the NMS12 countries are by far the most affected in quantitative terms, but they also exhibit the largest inequality levels, with 31% of the population lacking a flush toilet in the dwelling among the lowest income quintile and 3% in the highest income quintile.


Fig. 6. Prevalence of lack of a flush toilet for single-parent households versus general population (2009)

source?? Single parent household wirklich niedriger als tot pop?

EU27

NMS12

SVK

SVN

ROM

POL

LTULTV

HUN

EST

CZH

BUL

EU15

UNK

POR

LUX

ITAIRE

DEU

FRA

FIN

BEL

AUS

0.1

1

10

100

0 5 10 15 20 25 30 35 40 45

General population without flush toilet (%)

Sin

gle

pare

nt h

ouse

hold

s w

ithou

t flu

sh to

ilet (

%)

Source: data from EU-SILC, 2011.Note: see list of country abbreviations in Annex 4.

Fig. 7. Prevalence of lack of a flush toilet in the dwelling by income quintile (2009)

0

5

10

15

20

25

30

35

72UE21SMN51UE

Prev

alen

ce (%

)

Quintile 1 (lowest income)

Quintile 2

Quintile 3

Quintile 4

Quintile 5 (highest income)

Source: data from EU-SILC, 2011.

Target groups for actionThe analysis demonstrates that households in the NMS12 countries are the most exposed to lack of a flush toilet in the dwelling and most vulnerable in income-related terms. Moreover, it reveals greater inequalities between the socioeconomically advantaged and disadvantaged population groups in this


region; this corroborates the wealth of knowledge from the existing literature which identifies low-SES populations as having the poorest access to good sanitation conditions. Inequalities can also be observed among the EU15 countries, but to a lesser extent.

The income gradient analysis per region gives additional information underlining a strong linear decrease in prevalence of lack of a flush toilet in the dwelling from the lowest to the highest quintile. This reveals that inequalities exist not only between the poorest and the richest but also as a continuum across all income categories. The analysis also demonstrates that although when compared with the general population single-parent households are not disadvantaged by default, single-parent households in relative poverty have a prevalence of lack of a flush toilet almost double that for all single-parent households.

Health implicationsThe plethora of preventable health effects (such as acute diarrhoea – particularly among infants and children – hepatitis, typhoid and paratyphoid enteric fevers, intestinal parasitic worms and other parasitic diseases) related to lack of access to a flush toilet has been internationally recognized by the insertion of important water and sanitation targets in the Millennium Development Goals (United Nations, 2010). WHO has reiterated that progress towards these targets “will contribute significantly to the reduction of child mortality, major infectious diseases, maternal health and quality of life of slum populations” (Hutton and Bartram, 2008). Sanitation interventions have been shown to reduce illness and to have a substantial impact on the epidemiology of child diarrhoea: directly by reducing exposure to disease determinants, and indirectly by altering the pathways by which socioeconomic factors act on the outcome (Fewtrell et al., 2005; Genser et al., 2008).

Conclusions and suggestionsLack of a flush toilet in the dwelling is still an issue in many countries within the WHO European Region. The proportion of households with no flush toilet varies between subpopulations, with those in relative poverty – and particularly single-parent households in relative poverty – more exposed.

A few suggestions for action stem from this analysis and concern permanent individual and collective housing. Specific recommendations could also be formulated for temporary housing – such as that built for transitory workers or students’ residences – or any other provisional or emergency type of accommodation. The analysis does not take into account this particular category of housing.

Suggested mitigation actions are:• ensuring that all new residential buildings have at least one inside flush toilet per dwelling; • rehabilitation of existing dwellings to install a flush toilet where one is lacking;• offering financial support to disadvantaged populations to encourage installation of a flush

toilet in their housing;• better reporting of data for non-EU Member States, especially in relation to income and

household types not covered by WHO/UNICEF JMP data.

Evidently, the two first recommendations are only practical if households have access to an adequate water supply and if this supply is affordable for the poorest segments of the population.


INEquAlITIES IN lACK OF A bATH Or SHOwEr IN THE dwEllING

IntroductionThe absence of a bath or shower in the dwelling is associated with an inadequate supply of water and impairs personal and domestic hygiene. The Office of the High Commissioner for Human Rights (OHCHR) has stated that the right to water is intrinsically connected to the right to an adequate standard of living because it greatly influences human health and quality of life (OHCHR, 2011). Lack of a bath or shower in the dwelling severely undermines the capacity of individuals to use the required amount of water necessary to meet their most basic hygiene needs and to promote good health (Howard and Bartram, 2003; WHO, 2011c). Washing may not occur as often as necessary to maintain good hygiene and bathing cannot be assured unless it is carried out at an open water source outside the household, such as a nearby river, lake or stream. The water quality at these sources may even be contaminated and unsafe for human use. Moreover, if household members have to collect water to be used in the dwelling from outside, the distances and time involved may prevent the household from securing the volumes necessary to support optimal basic personal and domestic hygiene; this may also result in storage of the water at home, making it prone to contamination, even if it is initially clean (Howard and Bartram, 2003). Populations lacking a bath or shower in the dwelling, and consequently most at risk of contracting diseases related to suboptimal hygiene, are in addition those who are most socioeconomically disadvantaged and living in poverty. Even if they have access to minimum volumes of water from a source outside the dwelling, marginalized populations are unlikely to benefit from the protective effect of higher volumes of water per day, gained from water piped into the home (Luna et al., 1992; Pruss and Mariotti, 2000).

Indicator analysis: inequalities by income and household typeAs with the absence of a toilet in the dwelling indicator, the inequalities related to a lack of bath or shower in the dwelling are predominantly intercountry inequalities (see Fig. 8).

Fig. 8. Prevalence of lack of a bath or shower in the dwelling by country (2009)

0

5

10

15

20

25

30

35

40

45

Net

herla

nds

[a]

Spa

in [a

]Lu

xem

bour

g [b

]U

nite

d K

ingd

om [b

]G

erm

any

[b]

Italy

Sw

eden

Fran

ceIre

land

Aus

tria

Den

mar

kB

elgi

umFi

nlan

dG

reec

eP

ortu

gal

EU

15

Mal

ta [b

]S

lova

kia

[b]

Cze

ch R

epub

licS

love

nia

Cyp

rus

Hun

gary

Pol

and

Est

onia

Bul

garia

Lith

uani

aLa

tvia

Rom

ania

NM

S12

EU

27

Icel

and

[b]

Nor

way

[b]

Sw

itzer

land

[b]

Pre

vale

nce

(%)

Source: data from EU-SILC, 2011.Notes: [a] countries reporting full coverage; [b] countries reporting prevalence lower than or equal to 0.3%.

Having no bath or shower at home is not an issue for the EU15 countries, where the average prevalence is 0.4%. Portugal (as in the case of lack of a toilet in the dwelling) has the highest prevalence at 2.7%. In the Netherlands and Spain, the entire population is reported to have a bath or a shower in the dwelling and low prevalence levels (up to 0.3%) were observed for Luxembourg, Germany and United Kingdom, as well as for Iceland, Norway and Switzerland. By contrast, as seen with the previous housing indicator,


large differences were observed between the NMS12 countries, with a high prevalence in Romania (41.2%), Latvia (18.2%), Lithuania (15.9%) and Bulgaria (15.6%) and a low prevalence in Malta (0.2%), Slovakia (0.3%), the Czech Republic (0.5%) and Cyprus (0.7%).

The Eurostat data indicate strong inequalities in access to a bath or shower at home between populations in the lowest income quintile and those in the highest. Prevalence levels for income quintiles show a steep gradient in the NMS12 countries, especially in the lowest income quintile where 29.6% of the population report a lack of bath or shower, which is 11% higher than the second-lowest quintile (see Fig. 9). In the EU15 countries, despite the low prevalence of the problem overall, a social gradient also clearly exists (ranging from 1% in the lowest-income population to 0.1% in the highest). In summary terms, the data indicate that within the lowest-income population group, more than 7% of EU citizens do not have a bath or shower in their home. However, detailed data for Kyrgyzstan (see Annex 1) indicate that outside the EU the problems are much greater, but can only be identified through national databases.

Fig. 9. Prevalence of lack of a bath or shower by income quintile (2009)

0

5

10

15

20

25

30

35

72UE21SMN51UE

Pre

vale

nce

(%)

Quintile 1 (lowest income)Quintile 2Quintile 3Quintile 4Quintile 5 (highest income)


In the majority of countries, prevalence of lacking a bath or shower in the dwelling is higher among single-parent households than among all households with dependent children (see Fig. 10), except in France and Hungary. The ratio varies between 0.5:1 for France and 4:1 for Belgium or Germany, but to ensure a complete interpretation of the figures the low prevalence of households with no bath or shower in these countries (0.2%, 0.2% and 0.1% in all households with dependent children and 0.1%, 0.8% and 0.4% among single-parent households, respectively) must be noted.

Larger inequalities can be found when figures for single-parent households in relative poverty and for all households with dependent children are contrasted, with ratios ranging from 1.2:1 in Finland and about 12:1 in Denmark (see Fig. 10). This comparison does not reveal differences in the extent of inequalities between EU15 countries and NMS12 countries: the prevalence of living in a house without a bath or shower among low-income single-parent households is about twice greater than among households with dependent children across both regions.

In several countries, taking relative poverty into account only slightly increases the difference between figures for single-parent households and for all households with dependent children (see Fig. 10). This means that the issue concerns all single-parent households equally, irrespective of income. For instance, in Belgium both ratios are 4:1 and in Ireland both are 2:1. In contrast, Denmark and Slovenia exhibit a much larger difference when considering single-parent households in relative poverty: for both countries, the ratio is approximately 12:1. High ratios are also found for the Czech Republic and Poland.


Fig. 10. ratio of prevalence of lack of a bath or shower for all households with children compared to single-parent households, by poverty level (2009)

0

3

6

9

12

15

Gre

ece

[a]

Luxe

mbo

urg

[b]

Net

herla

nds

[b]

Spa

in [b

]S

wed

en [a

]U

nite

d K

ingd

om [a

]Fr

ance

Finl

and

Italy

Irela

ndP

ortu

gal

Bel

gium

Aus

tria

Ger

man

yD

enm

ark

EU

15

Cyp

rus

[a]

Mal

ta [a

]S

lova

kia

[a]

Hun

gary

Est

onia

Rom

ania

Latv

iaLi

thua

nia

Bul

garia

Cze

ch R

epub

licP

olan

dS

love

nia

NM

S12

EU

27

Nor

way

[a]

Sw

itzer

land

[a]

Icel

and

[b]

Rat

io

Ratio of prevalence (all single-parent households:all households with children)Ratio of prevalence (single-parent households below relative poverty level:all households with children)

Source: data from EU-SILC, 2011.Notes: [a] countries reporting full coverage with bath/shower for all single-parent households; [b] countries reporting full coverage with bath/shower for all households with dependent children.

Target groups for actionThe analysis confirms that prevalence of lack of a bath or shower in the dwelling is highest in a majority of NMS12 countries compared to EU15 countries: the average figures across each region are 13.4% and 0.4% respectively. Lower-income populations are also clearly identified as constituting a disproportionately highly exposed group. The current economic crisis with its serious social consequences might exacerbate existing environmental health risks, including those relating to lack of hygiene and sanitation equipment. On average among the EU15 countries, the prevalence of lacking a bath or shower among those below the relative poverty threshold is about four times higher than among those above it. The contrast is greater in the NMS12 countries, where the average ratio is 6:1. Finally, single-parent households – especially those on low incomes – are a particularly vulnerable group regarding lack of a bath or shower in the dwelling, compared to all households with dependent children.

Health implicationsA lack of hygiene and sanitation equipment is still one of the most basic threats to health found in dwellings (Bonnefoy, 2007). It has long been established that a number of diseases are linked to poor hygiene, including those transmitted through the faecal-oral route, skin and eye diseases, and diseases propagated by infestations, including louse- and tick-borne typhus (Bradley, 1977; Cairncross and Feachem, 1993). A study conducted in the south of Tehran found that lack of a bath or shower constitutes a risk factor for persistent diarrhoea in children under five years old (Sakhaie et al., 2001).

The main purpose of personal hygiene is to prevent such health-threatening conditions, but it is also relevant to improve appearance. This plays an important role in social acceptance and inclusiveness, since poor hygiene can hinder a person’s reputation, community integration and job capacity; factors that, in turn, are determinants of self-esteem and revenue.


Conclusions and suggestionsLack of a bath or shower is an issue in several countries, mainly among the NMS12 countries, and probably also among the countries in the eastern part of the WHO European Region, for which data were not available. Several subpopulations were identified as more frequently vulnerable to the absence of a bath or shower in the dwelling: socioeconomically disadvantaged groups, single-parent households and households combining these two sociodemographic determinants.

Suggested mitigation actions are:• ensuring that all new residential buildings have a bath or shower in each dwelling; • rehabilitation of existing buildings to install a bath or shower where one is lacking;• offering financial support to disadvantaged populations to encourage installation of a bath or

shower in their housing;• better reporting of data for non-EU member states, especially in relation to income and

household types not covered by WHO/UNICEF JMP data.

Evidently, the two first recommendations are only practical if households have access to an adequate water supply and if the cost of energy to heat the water is affordable for the poorest segments of the population.

INEquAlITIES IN OvErCrOwdING

IntroductionAn overcrowded house is recognized as an important contributor to ill health (Maani, Vaithianathan and Wolfe, 2006), a burden affecting several population subgroups. The 2008 annual report of the European Federation of National Organisations working with the Homeless (FEANTSA) advanced that among the population groups at greatest risk of overcrowded living are immigrants, refugees and ethnic minority groups; this is partly because overcrowded homes may be accepted as temporary accommodation by new arrivals to a city, particularly ethnic minorities (FEANTSA, 2008). For example, black and minority ethnic households in England are about six times as likely as white households to be overcrowded, and around half of the registered migrants in Austria live in an overcrowded home with at least two people per room. However, no data were identified that would enable an international inequality assessment of overcrowding in relation to migrant status or ethnicity.

The authors of a New Zealand study found that unemployed people are more likely to live in overcrowded households than people with full-time jobs, with respective proportions of 20% and 7% (Ministry of Social Development, 2010), and highlighted a clear relationship between income levels and overcrowding. Multigenerational households also logically constitute a group at particular risk of overcrowded living; for example, 83% and 67% of overcrowded households in Wales and in Scotland respectively include dependent children (FEANTSA, 2008). Moreover, it must be noted that there are also issues where elderly people cohabit with their children’s families – a situation more often seen in southern European countries.

Indicator analysis: inequalities by income and household typeThe problem of overcrowding is a particularly significant housing threat as it is suffered by 10.1% of the general population in the EU15 countries and 46.6% in the NMS12 countries, indicating strong subregional differences (see Fig. 11). It is unsurprising that overcrowding among households with


dependent children, which tend to be larger than the average household size, has a higher prevalence (14.4%) in the EU15 countries and a markedly higher prevalence (60.1%) in the NMS12 countries. Even higher prevalence levels are observed among single-parent households (EU15: 19.1%, NMS12: 68.2%).

Fig. 11. Prevalence of overcrowding by household type (2009)

0

10

20

30

40

50

60

70

80

90

Net

herla

nds

Spa

inIre

land

Bel

gium

Finl

and

Uni

ted

Kin

gdom

Den

mar

kP

ortu

gal

Fran

ceS

wed

enLu

xem

bour

gG

erm

any

Aus

tria

Italy

Gre

ece

EU

15

Cyp

rus

Mal

taC

zech

Rep

ublic

Bul

garia

Slo

vaki

aS

love

nia

Est

onia

Hun

gary

Pol

and

Rom

ania

Latv

iaLi

thua

nia

NM

S12

EU

27

Sw

itzer

land

Nor

way

Icel

and

Pre

vale

nce

(%)

All householdsHouseholds with dependent childrenSingle-parent households


Larger disparities in levels of inequality between household types can be observed among the EU15 countries than the NMS12 countries. In Austria and Belgium, the prevalence of living in an overcrowded house among single-parent households is at least double that among all households, while in Germany it is 3.5 times higher. Inequalities between household types are less apparent in the NMS12 countries, where overcrowding is more frequent – only the Czech Republic has a prevalence among single-parent households double that among all households. Nevertheless, protecting vulnerable groups in the NMS12 countries from overcrowding is a significant challenge, since for four countries the overcrowding prevalence among single-parent families exceeds 70%.

Fig. 11 also demonstrates an inverse correlation between the relative inequality and the absolute prevalence of overcrowding among households: the relative differences between household types are globally higher in the EU15 countries where total prevalence is lower, while the reverse is observed in the NMS12 countries.

In addition, a clear gradient of the prevalence of overcrowding appears in the general population according to income (see Fig. 12). For the EU15 countries, 20% of the lowest-income population are exposed to overcrowding versus less than 4% of the highest-income. For the NMS12 countries, a sharp decrease in prevalence was also found across the income quintiles, ranging from almost 60% to 35% between the first and the fifth quintiles. The data also indicate that the prevalence of overcrowding is much higher among single-parent households than the general population across all income quintiles, although it does not exactly follow the linear pattern observed in the general population: in the EU15 countries, overcrowding among single-parent households in the fourth income quintile (the second highest income category), is slightly greater than in the middle quintile, and there is also an inverse trend between the two highest income categories in the NMS12 countries.

Overall, the data confirm that the poorest quintile of the population suffers greater exposure to overcrowding than the richest quintile. This is confirmed by more detailed data from Great Britain (see Annex 1) showing that the overall risk of overcrowding also differs by dwelling tenure, but that household income remains associated with overcrowding in all types of tenure.


Fig. 12. Prevalence of overcrowding by income quintile and household type (2009)

0102030405060708090

Qui

ntile

1 (

low

est)

Qui

ntile

2

Qui

ntile

3

Qui

ntile

4

Qui

ntile

5 (h

ighe

st)

Qui

ntile

1 (

low

est)

Qui

ntile

2

Qui

ntile

3

Qui

ntile

4

Qui

ntile

5 (h

ighe

st)

Qui

ntile

1 (

low

est)

Qui

ntile

2

Qui

ntile

3

Qui

ntile

4

Qui

ntile

5 (h

ighe

st)

72UE21SMN51UE

Pre

vale

nce

(%)

General population

Single-parent households


Target groups for actionThe majority of NMS12 countries show high overcrowding figures compared to the EU15 countries. More detailed data show that several population subgroups are particularly exposed to overcrowding in both regions. Having dependent children in the household is a strong indicator for increased exposure to overcrowded living conditions, and single-parent households are at even greater risk.

Furthermore, a low income increases the risk of living in overcrowded homes. The prevalence of overcrowding is higher among low-income single-parent households (77%) than among all single-parent households (60%) in the NMS12 countries. These observations are in accord with the existing literature, which highlights the fact that low-income populations, and particularly single-parent households, are more susceptible to this form of discomfort and health risk. Analysis of the income gradient by subregion presents additional information underlining the gradual decrease in the prevalence of overcrowding from the lowest to the highest income quintile. This reveals that inequalities exist not only between the poorest and the richest households but also as a continuum across all income categories.

Health implicationsHousehold overcrowding is often associated with social deprivation, which in turn leads to an increased risk of infectious diseases (Maani, Vaithianathan and Wolfe, 2006). Numerous epidemiological studies have demonstrated the existence of a significant association between overcrowding and the prevalence of certain infectious diseases. Overcrowding may have a direct effect by facilitating the spread of infectious diseases such as tuberculosis, rheumatic fever and meningococcal disease (Baker et al., 2000). More specifically, tuberculosis associated with household overcrowding in the eastern part of the WHO European Region results in 0.8 deaths and 617.6 disability-adjusted life years (DALYs) per 100 000 population, corresponding to a total of 15 351 tuberculosis cases and 3518 deaths (Braubach, Jacobs and Ormandy, 2011).

Housing space adequate to the needs and desires of a family is also a core component of quality of life. Overcrowding has been associated with mental health problems (Braubach, Jacobs and Ormandy, 2011). A study conducted in north-west England found an association between overcrowding and the prevalence of psychiatric morbidity in the adult population (Harrison, Barrow and Creed, 1998). Several studies have also demonstrated that housing quality constitutes a good predictor of psychological issues and that overcrowding in particular is significantly associated with children’s mental health (Evans, Saegert and Harris, 2001; Evans, Saltzman and Cooperman, 2001; Evans and English, 2002).


Conclusions and suggestionsOvercrowding is an issue in the majority of NMS12 countries and, to a lesser extent, in the EU15 countries. Several subgroups were identified as more frequently exposed to overcrowding in the dwelling, although by far the greatest overcrowding risk is found for socioeconomically disadvantaged households. Households with children in general and specifically single-parent households also show higher prevalence rates. Those accumulating both sociodemographic determinants have the highest exposure levels: low-income single-parent households in the NMS12 countries have a prevalence of overcrowding of more than 75%.

Suggested mitigation actions are:• making regulatory provisions so that all new residential building projects, private or public,

plan a minimum proportion of dwellings for large households – this should take into account the demographic characteristics and future trends of each country and, where appropriate, of different regions within a country;

• ensuring a significant proportion of large dwellings when renovating existing housing stock;• providing targeted financial support to disadvantaged populations to facilitate access to

dwellings of sufficient size;• increasing public housing programmes which provide dwellings with an adequate number of

rooms as a way to improve the overcrowding situation of low-income populations and those groups most vulnerable to overcrowding.

INEquAlITIES IN dAmPNESS IN THE HOmE

IntroductionDampness in the dwelling constitutes a substandard living condition that indicates the presence of water damage, a leaking roof, rot in window frames and floors, visible mould or condensation (Social Care Institute for Excellence, 2005). Dampness is associated with a broad array of detrimental health effects in adults and children (Fisk, Lei-Gomez and Mendell, 2007). Mould growth is also facilitated when dampness is present indoors, a condition often encountered in poor social conditions associated with large family size that gives rise to humidity (Butler et al., 2003). Individuals with lower SES are highly exposed to the effects of substandard housing conditions: studies have shown that ethnic minorities and individuals with low income tend to suffer disproportionately from the adverse health effects related to exposure to dampness and mould (Kohlhuber et al., 2006; Bolte and Mielck, 2004; Mielck and Heinrich, 2002). A WHO report on guidelines for indoor air pollution concludes that dampness should be given greater priority as it contributes to the deterioration in health of disadvantaged and low-income populations who are already overburdened by disease (WHO, 2009).

Indicator analysis: inequalities by income and household typeA comparable level of prevalence of dampness in the dwelling exists in both the EU15 and the NMS12 countries (see Fig. 13).

On average, 15% of the general population is affected by dampness in the home in the EU15 versus 18% in the NMS12 countries. However, within these regional averages, strong national variations are observed. The lowest prevalence is found in Finland, where only 5% of the population live in damp homes; similarly low levels were found in Sweden and Slovakia. Slovenia has the highest prevalence at 30%, followed by Cyprus at 29%.


Fig. 13. Prevalence of dampness in the dwelling by household type (2009)

0

5

10

15

20

25

30

35

40

45

Finl

and

Sw

eden

Den

mar

kG

reec

eIre

land

Aus

tria

Por

tuga

lS

pain

Fran

ceU

nite

d K

ingd

omLu

xem

bour

gN

ethe

rland

sIta

lyB

elgi

umG

erm

any

EU

15

Slo

vaki

aM

alta

Cze

ch R

epub

licH

unga

ryB

ulga

riaE

ston

iaLi

thua

nia

Latv

iaP

olan

dR

oman

iaS

love

nia

Cyp

rus

NM

S12

EU

27

Nor

way

Sw

itzer

land

Icel

and

Pre

vale

nce

(%)

0

1

2

3

Rat

io o

f pre

vale

nce

(sin

gle

pare

nt:a

ll ho

useh

olds

)

All householdsSingle-parent householdsHousehold ratio


Moreover, in the majority of countries the prevalence of dampness in housing is higher among single-parent households than the general population (household ratio of 1.5:1 for EU27, with national household ratios ranging from 0.9:1 to 2.1:1). The prevalence of dampness in single-parent households across regions is 24% in the EU15 and 25% in the NMS12 countries, but again there are wide national variations, with the highest prevalence of 41% in Cyprus, followed by 39% in Slovenia, and the lowest once again in Finland at 7%. The prevalence of dampness among single-parent households in Germany is double that among the general population at 29% and 14% respectively – the largest difference measured. However, differences of 10% or more are also found between the general population and single-parent households in Belgium, Cyprus, France, the Netherlands, Poland and Romania. In contrast, Greece and Bulgaria report the opposite situation, with a slightly higher prevalence of damp homes in the general population than among single-parent households. More details on the association of household types with dampness can be found in data from Norway (see Annex 1).There is a clear gradient in the prevalence of dampness in homes from the lowest income quintile to the highest (see Fig. 14). The trend is steeper among the NMS12 than the EU15 countries, a result partly explained by the high average prevalence in the lowest income quintile among the NMS12 countries (30%).

Fig. 14. Prevalence of damp dwellings by income quintile (2009)

0

5

10

15

20

25

30

35

72UE21SMN51UE

Pre

vale

nce

(%)




A more in-depth analysis by country (data not shown) highlights wide income-related variations, particularly among EU15 countries. Southern European countries such as Portugal, Greece and Italy show the highest prevalences in the lowest income quintiles (between 25% and 28%) while in Nordic countries (such as Denmark, Finland, Norway and Sweden), the prevalence levels are much lower for these groups (between 7% and 12%), leading to lower levels of inequality. Among the NMS12 countries, the steepest gradient of dampness across the income quintiles is found in Romania and Poland, with 35.5% and 30.1% prevalence in the lowest income group and 8.7% and 7.4% in the highest, respectively. Compared to this, countries such as Slovenia (40.3% versus 21.4%) and Cyprus (35.0% versus 21.2%) show a lower level of inequality, but this is mostly because problems with damp are also common for high-income households.

Target groups for actionIn contrast to the other housing inequality indicators considered above, living in a damp dwelling is a housing issue in both the EU15 and the NMS12 countries. Lower-income populations in both regions (especially in the lowest income quintile) constitute more exposed groups. This is particularly marked among the NMS12 countries, where a large gap exists between the lowest income quintile and the others, giving a three-fold difference in exposure to dampness in the home between poor and rich. Single-parent households are also a group at higher risk: in all countries bar two, the risk of living in a damp dwelling is higher for single-parent households than among the general population.

Health implicationsDampness is associated with a broad array of detrimental health effects; the most common of these are related to the deterioration of the respiratory system (Mudarri and Fisk, 2007), including a higher prevalence of respiratory symptoms, increased risk of asthma, wheezing cough (Pirastu et al., 2009) bronchitis, common cold and rhinitis (Pirhonen et al., 1996). Some studies showed a clear relation between dampness and mould and objective measures of lung function. Children’s respiratory health is particularly damaged by living in a damp or mouldy home (Tischer et al., 2011). Results based on the analysis of data collected from 45 countries of the WHO European Region estimate that a considerable proportion of childhood asthma cases are attributable to exposure to indoor dampness and mould (Braubach, Jacobs and Ormandy, 2011). More precisely, it has been estimated that 0.07 asthma-related deaths and 50 asthma-related DALYs per 100 000 children per year are associated with exposure to dampness in dwellings, and that 0.06 asthma-related deaths and 40 asthma-related DALYs per 100 000 children per year are associated with exposure to mould. Irritations of the throat and eyes, allergies, rhino-conjunctivitis and eczema have also been observed repeatedly (Bonnefoy, 2007; Zacharasiewicz et al., 2000; Simoni et al., 2005; McNally, Williams and Phillips, 2001).

Conclusions and suggestionsDampness in the home is an issue in all 27 European countries for which data were available. A large proportion of the general population lives in homes where dampness is an issue. As with many other risk factors, socioeconomically disadvantaged households (particularly those in the two lowest income quartiles) are most affected, as are single-parent households.

One reason for dampness and associated mould is related to the design and construction of buildings; good design and proper construction can help to prevent problems from occurring (WHO, 2010b). Maintenance and use of buildings can also be considered key factors to preserve healthy housing; for example, a speedy response to water damage will help to keep the dwelling in sound condition.


Suggested mitigation actions are:• making regulatory provisions so that all new residential building projects, private or public,

provide adequate protection of building structures against excess dampness and humidity – adequate levels of ventilation should be maintained in all buildings, irrespective of their use and ownership;

• ensuring that the problems of dampness, humidity and ventilation are considered when renovating existing and especially low-cost housing stock;

• providing targeted financial support to disadvantaged populations and those groups most exposed to damp homes due to specific housing circumstances;

• providing adequate housing conditions and affordable heating to economically disadvantaged and vulnerable larger households, since overcrowded living conditions and indoor cold both contribute to dampness.

The last recommendation links the problem of dampness to other housing inequality indicators discussed in this chapter: policies and actions to reduce overcrowding and increase thermal comfort will have an indirect impact on the prevalence of dampness in the homes of disadvantaged populations groups.

INEquAlITIES IN KEEPING THE HOmE AdEquATEly wArm

IntroductionIn colder climates, living in a comfortably heated home is commonly viewed as protective of human health. WHO recommends a minimum temperature of 21°C in living rooms, and 18°C in all other rooms (WHO, 2007). Households unable to maintain this standard comfortable temperature, or that require more than 10% of their income to attain the WHO standards, are described as living in fuel poverty (European fuel poverty and energy efficiency: Epee Project, 2008). Many countries have policies that protect their most vulnerable population from cold-related health risks; the United Kingdom Fuel Poverty Strategy (BERR, 2001), for example, aims to reduce fuel poverty by focusing on people over 60 years old, people living with disabilities or long-term illnesses, and households with dependent children, and subsequently to eradicate fuel poverty by 2018 (Liddell and Morris, 2010).

Fuel poverty disproportionately affects low-income households, which must economize across all areas of the household budget and choose the most important expenses for their family. This economic disadvantage is exacerbated by the poor energy efficiency standards of many of their homes. In Europe, between 50 and 125 million people are estimated to suffer from fuel poverty and the Epee Project reveals that this number will inevitably rise in the future as global energy prices increase. Approximately one-third of households in Northern Ireland and Wales live in fuel poverty, whereas 13% in Scotland and 9% in England are affected by this problem (Shortt and Rugkåsa, 2007).

It is important to recognize that a household’s inability to keep the home adequately warm has serious health consequences. Extreme winter temperatures have been demonstrated to cause the premature deaths of tens of thousands of EU citizens each year. In a European cross-country analysis (Healy, 2003) a significant association between poor housing – in terms of thermal efficiency – and high levels of winter mortality was found. Further evidence suggests that the excess of winter mortality in the United Kingdom is related to fuel poverty and is highest among the most disadvantaged population groups (Braubach and Fairburn, 2010).


Indicator analysis: inequalities by income, Gini index and household typeInability to keep the home warm constitutes a housing issue among both the NMS12 countries (18.4% prevalence among the general population) and – although to a lesser extent – the EU15 countries (6.9% prevalence). Of the NMS12 countries, Bulgaria has by far the highest prevalence at 64.2%, while Estonia has the lowest at 1.7%; of the EU15 countries, Portugal has the highest prevalence and Luxembourg the lowest, at 28% and 0.3% respectively.

Income levels influence the risk of exposure to cold homes across both subregions, and for many countries, households below the poverty threshold are particularly strongly affected. The magnitude of income-related inequalities is in fact greater among the EU15 than among the NMS12 countries: on average, the prevalence of inability to keep the home warm is 3.4 times higher for those in relative poverty in the EU15 countries, but only 2.1 times higher in the NMS12 countries (see Fig. 15).

Fig. 15. Prevalence of inability to keep the home warm by relative poverty level (2009)

Last bracket on second y-axis missing

0

10

20

30

40

50

60

70

80

90

Luxe

mbo

urg

Den

mar

kFi

nlan

dS

wed

enN

ethe

rland

sA

ustri

aIre

land

Uni

ted

Kin

gdom

Spa

inB

elgi

umFr

ance

Ger

man

yIta

lyG

reec

eP

ortu

gal

EU

15

Est

onia

Slo

veni

aS

lova

kia

Cze

ch R

epub

licM

alta

Hun

gary

Latv

iaR

oman

iaLi

thua

nia

Pol

and

Cyp

rus

Bul

garia

NM

S12

EU

27

Icel

and

Nor

way

Sw

itzer

land

Pre

vale

nce

(%)

0

1

2

3

4

5

6

7

8

9

Rat

io o

f pre

vale

nce

(bel

ow:a

bove

rela

tive

pove

rty le

vel)

Above relative poverty levelBelow relative poverty levelIncome ratio


There is a general trend that countries with high prevalence rates of inability to keep the home warm have lower income-related inequalities. The inequality ratio of 1.3:1 in Bulgaria is the lowest among the NMS12 countries, which, combined with the country’s high prevalence rate, shows that keeping the home warm represents a widespread housing problem. The highest ratio is found in Luxembourg, at 5.5:1, but evaluation of this relative metric must take two facts into account: only a small proportion of the general population (about 0.3%) reported the problem, while prevalence among households below the poverty threshold is 1.1%. Similar findings are seen for Austria, the Netherlands, Estonia, Slovakia and Norway, all of which present a high level of inequality between populations above and below the poverty threshold (with a ratio of about 4:1 or higher), but a very low prevalence in the general population, at less than 4%. In comparison, countries such as Cyprus, Greece, Latvia and Poland may need to pay particular attention to this indicator, since their inequality ratios are at least 2:1 and high prevalence rates (above 15%) are observed in the general population.

There is also a clear linear association between the Gini index indicating national income-related inequalities and reported inability to keep the home warm (see Fig. 16): about 16% of the variability of the inequalities between the population above and below the poverty threshold is explained by the Gini index (R2=0.16). When the index increases by 1%, income inequalities related to thermal comfort in winter decrease by 0.12. In other words, a lower Gini value, indicating a lower general income inequality,


goes with a higher income inequality related to thermal comfort. For example, comparing Norway with Portugal, income inequalities are higher in Portugal (Gini index of 24.1 in Norway and 35.4 in Portugal) whereas inequalities between populations above and below the poverty threshold are higher in Norway (ratios of 4:1 in Norway and 1.8:1 in Portugal).

Fig. 16. ratio of prevalence of inability to keep the home warm (households below:households above relative poverty level, 2009) in relation to the Gini index (2009)

country acronym alloca�on to be checked a�er impot into text chapter

AUTBEL

DEN

FIN FRA

DEU

GREIRE ITA

LUX

NET

POR

SPA

SWE

UNK

EU15

BUL

CYPCZH

EST

HUNLVA

LTUMAT

POL

ROM

SVN

SVK

NMS12

EU27

ICE

NOR

SWI

0

1

2

3

4

5

6

20 22 24 26 28 30 32 34 36 38 40

Gini index

Pov

erty

ratio

Source: data from EU-SILC, 2011.Note: see list of country abbreviations in Annex 4.

This means that inability to keep the home warm is inversely related to the general level of income inequality within countries, indicating that inequalities in thermal comfort in winter are not directly associated with national income inequality patterns. This effect may result from the fact that in some countries the difference of prevalence between income groups is rather low (see Fig. 15).

The demographic groups most exposed to the problem of inability to keep the home warm are not the same across European countries (see Fig. 17). One common point is that households with dependent children seem less affected by the problem (7.1% in the EU15 and 17.4% in the NMS12 countries) than single households with one adult older than 65 years (8.3% in the EU15 and 27.5% in the NMS12 countries) and single-parent households (13.1% in the EU15 and 25.1% in the NMS12 countries). By contrast, single-parent households have greater difficulty keeping the home warm among the EU15 countries, while among the NMS12 countries, the overall prevalence is higher for single households with one adult older than 65 years. Among the EU15 countries, Portugal is an exception: the prevalence of inability to keep the home warm is much higher among single households with one adult older than 65 years (44%) than among single-parent households (31%).

Target groups for actionMost of the NMS12 countries present a high prevalence of problems with keeping the home warm in the general population, with Bulgaria showing by far the highest level. Within countries, socioeconomically deprived populations report more problems with thermal comfort in winter in both the EU15 and the NMS12 countries, although the magnitude of this inequality is greater in western Europe. This result is consistent with results obtained for the five other housing indicators and corroborates the wealth of knowledge in the existing literature reporting that socioeconomically disadvantaged groups are more susceptible to environmental exposures.


Fig. 17. Prevalence of inability to keep the home warm by household type (2009)

0

10

20

30

40

50

60

70

80Lu

xem

bour

gFi

nlan

dS

wed

enN

ethe

rland

sD

enm

ark

Aus

tria

Uni

ted

Kin

gdom

Fran

ceIre

land

Spa

inB

elgi

um Italy

Ger

man

yP

ortu

gal

Gre

ece

EU

15

Est

onia

Slo

vaki

aS

love

nia

Cze

ch R

epub

licH

unga

ryM

alta

Latv

iaC

ypru

sR

oman

iaP

olan

dLi

thua

nia

Bul

garia

NM

S12

EU

27

Icel

and

Nor

way

Sw

itzer

land

Pre

vale

nce

(%)

Households with dependent childrenSingle-parent households with dependent childrenHouseholds with one adult older than 65 years


In the majority of countries, single-parent households are most frequently unable to keep the home warm and should be considered a group at high risk. Households with one adult older than 65 also constitute a group prone to the problem in a few countries, including Bulgaria, Cyprus and Portugal. Further detailed information on potentially vulnerable groups is available from Serbia and Montenegro, showing, for example, that education level may also have a strong impact on ability to heat the home (see Annex 1).

Health implicationsAccording to the recent Marmot Review Team report (2011) on the health impacts of cold housing and fuel poverty, excess winter deaths (EWDs) are associated with thermal efficiency of housing and low indoor temperatures. About 40% and 33% of EWDs are attributable to cardiovascular and respiratory diseases respectively, the risk of EWD being almost three times higher in the quartile of houses with coldest indoor temperatures than in the warmest quartile. A WHO report on the burden of disease of inadequate housing quantified as 30% the proportion of EWDs attributable to cold housing (Braubach, Jacobs and Ormandy, 2011). A BMJ editorial commenting on the Marmot Review Team report concludes that “although the extra deaths in elderly people are caused mainly by cardiovascular and respiratory disease, far greater numbers have minor ailments that lead to a huge burden of disease, costs to the health system, and misery” (Dear and McMichael, 2011). Non-fatal cardiovascular and respiratory diseases are also linked to low indoor temperature, which exacerbates existing conditions such as arthritis and rheumatism, increased blood pressure and risk of stroke, social isolation and adverse effects on children’s education and nutrition (Shortt and Rugkåsa, 2007). Children are especially at risk, in particular for respiratory problems that are, in part, due to the development of moulds in the humid environments of cold houses. The Marmot Review Team (2011) also reports that mental health is negatively affected by fuel poverty and cold housing in all age groups, with a fivefold increase in risk of poor mental health among adolescents living in cold housing than those in warmer houses.

Conclusions and suggestionsInability to keep the home warm is an important housing issue in most European countries, and a high level of inequality is also apparent. Population subgroups living in relative poverty are significantly more affected, with several countries showing three or four times higher prevalence rates for households below than those above the poverty threshold. Inequalities in exposure to cold homes are also identified for single-parent households and, to a lesser extent, households with one adult older than 65 years.


Suggested mitigation actions are:• offering special energy prices or financial support to the most vulnerable households (especially

those in relative poverty and households with children), allowing them to pay for the energy needed to maintain an appropriate temperature inside the home;

• ensuring efficient insulation to reduce energy consumption and related expenses – poorer households often live in the least energy-efficient homes, and will thus need financial support to afford thermal insulation;

• building and rehabilitation of social housing to meet energy standards and reduce households’ energy expenses;

• giving consideration at a local scale, in accordance with general energy-saving policies, to the availability of low-cost energy sources such as wood, residual agricultural biomass, wind and geothermal sources, depending on the local situation and opportunities;

• providing measures and information from a health perspective to avoid an extreme focus on saving energy leading to reduced air exchange and a deterioration of indoor air quality: a good balance is required between ventilation and insulation.

INEquAlITIES IN KEEPING THE HOmE AdEquATEly COOlAlthough it falls outside the environmental health inequality indicator set, the EU-SILC database also offers data on dwellings that cannot be kept cool in summer, enabling comparison with the opposite thermal comfort indicator discussed above. Globally, the proportion of the general population unable to keep the dwelling comfortably cool in summer is higher than the proportion unable to keep the home warm in winter, showing that summer temperatures may be a rising problem. Much higher prevalence levels can be found among the NMS12 countries (average 37.7%) than among the EU15 countries (average 24.2%).

Moreover, there is a clear gradient in the proportion of the population not comfortably cool in the dwelling during the summer from the lowest income quintile to the highest (see Fig. 18). The trend appears steeper among the EU15 countries than among the NMS12 countries, where the highest income quintile is much more affected than the lowest in the EU15.

Fig. 18. Prevalence of inability to keep the home adequately cool in summer by income quintile (2007)

0

5

10

15

20

25

30

35

40

45

50

72UE21SMN51UE

Pre

vale

nce

(%)




There are strong variations within both the EU15 and the NMS12 countries, with the overall prevalence of not being comfortably cool in summer largely affected by geographical location and climate (see Fig. 19). However, the data also indicate that the impact of income on problems with indoor temperatures in summer is very diverse, depending on the country.

The inequalities across the income quintiles seem most strongly expressed in hot climates such as Greece, Italy, Portugal and Cyprus, which show a prevalence difference of around or even beyond 20%, with Greece and Cyprus indicating an income ratio of 2:1 between the lowest- and highest-income population quintiles. Of further note is the finding in Lithuania and Hungary, where further exploration is needed to explain the reverse finding that high-income households are more affected. Finally, the data also indicate that the problem of keeping the home cool in summer is less clearly related to income levels than was the case for keeping the home warm in winter (where income ratios were higher than 3:1 for many EU15 and several NMS12 countries: see Fig. 15).

Fig. 19. Prevalence of inability to keep the home adequately cool in summer by income (2007)

0

10

20

30

40

50

60

Irela

ndU

nite

d K

ingd

omSw

eden

Belg

ium

Finl

and

Den

mar

kA

ustri

aN

ethe

rland

sG

erm

any

Luxe

mbo

urg

Fran

ceSp

ain

Gre

ece

Italy

Por

tuga

lEU

15

Mal

taE

ston

iaLi

thua

nia

Slov

enia

Hun

gary

Slov

akia

Cze

ch R

epub

licLa

tvia

Pola

ndC

ypru

sN

MS

12 c

ount

ries

EU

27 c

ount

ries

Icel

and

Nor

way

Pre

vale

nce

(%)

Quintile 1 (lowest income)Quintile 5 (highest income)All incomes


CONCluSION ON HOuSING-rElATEd INEquAlITIESFrom a public health perspective, two aspects of housing inequality can be distinguished: inequalities relating to the basic needs of water and equipment availability for drinking, cooking and hygiene on the one hand, and inequalities relating to overcrowding, dampness and the capacity to keep the home warm or cool on the other. The main health effects associated with water and sanitation indicators are infectious diseases, while those associated with the latter indicators are allergic and respiratory or cardiovascular diseases. This analysis reflects the distinction between these two different aspects, showing that the former is of particular concern in the NMS12 countries, whereas the latter affects both the NMS12 and the EU15 countries.

The analysis also clearly demonstrates that a high proportion of the population living in the NMS12 countries has no flush toilet and no bath or shower in the dwelling. Because absence of sanitation equipment inside the home has such considerable health consequences, housing policies should ensure that all new residences in public buildings or private houses offer such basic commodities. This requires that no construction should be allowed without a permit from local authorities that will check that not only the building plans but also the reality conform to this obligation. Regarding pre-existing premises, the data clearly show that poverty is the key risk factor for absence of a toilet and shower or bath.


Housing and public health policies should therefore target low-income household groups including single-parent families – a particularly vulnerable category – for the delivery of social aid and funds to help reduce the prevalence of this problem. Indeed, the study reveals that the combination of these two socioeconomic characteristics of poverty and single parenthood gives rise to a particularly exposed and vulnerable group that could be considered a priority for targeted action in many countries.

Accessibility (and affordability) of drinking-water has also been shown to be a serious issue in many countries of the eastern part of the WHO European Region – particularly, but not solely, in rural areas. Moreover, the analysis reveals that although the prevalence of inadequate water supply has tended to decrease for all countries since 1995, it remains very high for the Euro 3 countries. Taking the significant health consequences into account once again, the same rationale for the necessity for installing sanitation equipment should apply to the obligation to connect water networks to all new and old dwellings in urban areas. In addition, provision should be made at national and regional levels to help investments in rural areas that would develop community water systems in small towns and villages.

The data provide strong evidence that the non-sanitary housing inequalities – overcrowding, dampness and thermal comfort related to cool and warm homes – exist in almost every country. Once again, low-income populations, and especially low-income single-parent households, are most affected across all indicators. This requires particular attention because this combination of disadvantages has been identified as the group most exposed to substandard housing in many countries.

Progress in the areas of overcrowding, dampness and thermal comfort can be made through a combination of housing and social policies and should be regulated by national laws to ensure their effective implementation throughout the country. All new building programmes should be obliged to provide a minimum proportion of dwellings with a large number of rooms in order to host large families, with low-cost homes for rent or for purchase, the figures depending on national demographic and sociocultural characteristics. Rehabilitation programmes for existing buildings should aim to achieve the same goal. Dedicated financial support should also be targeted to economically disadvantaged groups to cope with the costs of energy (for heating or cooling) and to help them improve insulation and ventilation of the dwelling.

rEFErENCES

Baker M et al. (2000). Household crowding a major risk factor for epidemic meningococcal disease in Auckland children. Pediatric Infectious Disease Journal, 19(10):983–990.

Bernstein JA et al. (2008). The health effects of non-industrial indoor air pollution. Journal of Allergy and Clinical Immunology, 121(3):585–591.

BERR (2001).The UK fuel poverty strategy. London, Department for Business Enterprise and Regulatory Reform (http://webarchive.nationalarchives.gov.uk/+/http://www.berr.gov.uk/files/file16495.pdf, accessed 14 November 2011).

Bolte G, Mielck A, eds. (2004). Umweltgerechtigkeit. Die soziale Verteilung von Umweltbe-lastungen [Environmental justice. The social distribution of environmental exposures]. Weinheim, Juventa.

Bonnefoy X (2007). Inadequate housing and health: an overview. International Journal of Environment and Pollution, 30(3/4):411–429.

Bradley D (1977). Health aspects of water supplies in tropical countries. In: Feachem R, McGarry M, Mara D, eds. Water, wastes and health in hot climates. Chichester, John Wiley.

Braubach M, Fairburn J (2010). Social inequities in environmental risks associated with housing and residential location—a review of evidence. European Journal of Public Health, 20(1):36–42.

Braubach M, Jacobs D, Ormandy D, eds. (2011). Environmental burden of disease associated with inadequate housing: methods for quantifying health impacts of selected housing risks in the WHO European Region. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0003/142077/e95004.pdf, accessed 7 November 2011).


Breysse P et al. (2004). The relationship between housing and health: children at risk. Environmental Health Perspectives, 112:1583–1588.

Butler S et al. (2003). Problems with damp and cold housing among Pacific families in New Zealand. New Zealand Medical Journal, 116(1177):494.

Cairncross S, Feachem R (1993). Environmental health engineering in the tropics: an introductory text, 2nd ed. Chichester, John Wiley.

Clasen T et al. (2007). Interventions to improve water quality for preventing diarrhoea: systematic review and meta-analysis. British Medical Journal, 334(7597):782–785.

Dear KBG, McMichael AJ (2011). The health impacts of cold homes and fuel poverty. Three reasons to act: the health burden, inequity, and mitigation. British Medical Journal, 342:d2807.

European fuel poverty and energy efficiency: Epee Project [web site] (2008). Valbonne, Alpheeis (http://www.fuel-poverty.org, accessed 13 October 2011).

EU-SILC [online database] (2011). Luxembourg, Eurostat (http://epp.eurostat.ec.europa.eu/portal/page/portal/microdata/eu_silc, downloaded on 9 November 2011).

Evans GW, English K (2002). The environment of poverty, multiple stressor exposure, psychophysical stress, and socioemotional adjustment. Child Development, 73(4):1238–1248.

Evans GW, Saegert S, Harris R (2001). Residential density and psychological health among children in low income families. Environment and Behavior, 33(2):165–180.

Evans GW, Saltzman H, Cooperman J (2001). Housing quality and children’s socioemotional health, Environment and Behavior, 33(3):389–399.

FEANTSA (2008). European report: the role of housing in pathways into and out of homelessness. Annual theme 2008: housing and homelessness. Brussels, European Federation of National Organisations Working with the Homeless.

Fewtrell L et al. (2005). Water, sanitation, and hygiene interventions to reduce diarrhoea in less developed countries: A systematic review and meta-analysis. Lancet Infectious Diseases, 5(1):42–52.

Fisk WJ, Lei-Gomez Q, Mendell MJ (2007). Meta-analyses of the associations of respiratory health effects with dampness and mold in homes. Indoor Air, 17(4):284–296.

Galobardes B et al. (2006). Indicators of socioeconomic position (part 1). Journal of Epidemiology and Community Health, 60(1):7–12.

Genser B et al. (2008). Impact of a city-wide sanitation intervention in a large urban centre on social, environmental and behavioural determinants of childhood diarrhoea: analysis of two cohort studies. International Journal of Epidemiology, 37(4):831–840.

Harrison J, Barrow S, Creed F (1998). Mental health in the north west region of England: associations with deprivation. Social Psychiatry and Psychiatric Epidemiology, 33(3):124–128.

Healy JD. (2003) Excess winter mortality in Europe: a cross country analysis identifying key risk factors. Journal of Epidemiology and Community Health, 57(10):784–789.

Howard G, Bartram J (2003). Domestic water quantity, service level and health. Geneva, World Health Organization (http://www.who.int/water_sanitation_health/diseases/WSH03.02.pdf, accessed 7 November 2011).

Howden-Chapman P (2002). Housing and inequalities in health. Journal of Epidemiology and Community Health, 56:645–646.

Humphrey, JH (2009). Child undernutrition, tropical enteropathy, toilets, and handwashing. Lancet, 374(9694):1032–1035.

Hunter PR, Zmirou-Navier D, Hartemann P (2009). Estimating the impact on health of poor reliability of drinking water interventions in developing countries. Science of the Total Environment, 407(8):2621–2624.

Hutton G, Bartram, J (2008). Global costs of attaining the Millennium Development Goals. Bulletin of the World Health Organization, 86(1):13–19 (http://www.who.int/bulletin/volumes/86/1/07-046045/en/, accessed 24 2011).

Katukiza A et al. (2010). Selection of sustainable sanitation technologies for urban slums—a case of Bwaise III in Kampala, Uganda. Science of the Total Environment, 409(1):52–62.


Kohlhuber M et al. (2006). Social inequality in perceived environmental exposures in relation to housing conditions in Germany. Environmental Research, 101(2):246–255.

Krieger J, Higgins DL (2002). Housing and health: time again for public health action. American Journal of Public Health, 92(5):758–768.

Liddell C, Morris C (2010). Fuel poverty and human health: a review of recent evidence. Energy Policy, 38:2987–2997.

Luna EJ et al. (1992). Epidemiology of trachoma in Bebedouro State of Sao Paulo, Brazil: prevalence and risk factors. International Journal of Epidemiology, 21(1):169–177.

Maani SA, Vaithianathan R, Wolfe B (2006). Inequality and health: is housing crowding the link? Motu Working Paper 06–09. Wellington, Motu Economic and Public Policy Research.

Marmot Review Team (2011). The health impacts of cold homes and fuel poverty. London, Friends of the Earth and Marmot Review Team (http://www.foe.co.uk/resource/reports/cold_homes_health.pdf, accessed 7 November 2011).

McNally NJ, Williams HC, Phillips DR (2001). Atopic eczema and the home environment. British Journal of Dermatology, 145(5):730–736.

Mielck A, Heinrich J (2002). Social inequalities and distribution of the environmental burden among the population (environmental justice). Gesundheitswesen, 64(7):405–16.

Ministry of Social Development (2010). The social report 2010: te purongo oranga tangata. Wellington, Ministry of Social Development.

Mudarri D, Fisk WJ (2007). Public health and economic impact of dampness and mold. Indoor Air, 17(3):226–235.

OECD.Stat Extracts [online database] (2011). Paris, OECD (http://stats.oecd.org/Index.aspx, accessed 9 November 2011).

OHCHR (2011). Special Rapporteur on the human right to safe drinking water and sanitation [web site]. Geneva, Office of the High Commissioner for Human Rights. (http://www.ohchr.org/EN/Issues/WaterAndSanitation/SRWater/Pages/SRWaterIndex.aspx, accessed 24 May 2011).

Pirastu R et al. (2009). Indoor exposure to environmental tobacco smoke and dampness: respiratory symptoms in Sardinian children—DRIAS study. Environmental Research, 109(1):59–65.

Pirhonen I et al. (1996). Home dampness, moulds and their influence on respiratory infections and symptoms in adults in Finland. European Respiratory Journal, 9(12):2618–2622.

Poortinga W, Dunstan FD, Fone DL (2008). Neighbourhood deprivation and self-rated health: the role of perceptions of the neighbourhood and of housing problems. Health & Place, 14(3):562–575.

Pruss A, Mariotti SP (2000). Preventing trachoma through environmental sanitation: a review of the evidence base. Bulletin of the World Health Organization, 78(2):258–266.

Rauh VA, Landrigan PJ, Claudio L (2008). Housing and health: intersection of poverty and environmental exposures. Annals of the New York Academy of Sciences, 1136:276–288.

Sakhaie E et al. (2001). Risk factors for persistent diarrhea in children under five years of age in the south of Tehran. Iranian South Medical Journal, 3(2):117–122.

Sandel M, Wright RJ (2006). When home is where the stress is: expanding the dimensions of housing that influence asthma morbidity. Archives of Disease in Childhood, 91(11):942–948.

Shortt N, Rugkåsa J (2007). “The walls were so damp and cold” fuel poverty and ill health in Northern Ireland: results from a housing intervention. Health & Place, 13(1):99–110.

Simoni M et al. (2005). Mould/dampness exposure at home is associated with respiratory disorders in Italian children and adolescents: the SIDRIA-2 Study. Occupational and Environmental Medicine, 62(9):616–622.

Social Care Institute for Excellence (2005). SCIE research briefing 19: What is the impact of environmental housing conditions on the health and well-being of children? London, Social Care Institute for Excellence.

Tischer C et al. (2011). Respiratory health in children, and indoor exposure to (1,3)-β-D-glucan, EPS mould components and endotoxin. European Respiratory Journal, 37(5):1050–1059.


UNDP (2011). World water day: situation critical for water supply and sanitation in Europe and Central Asia. New York, UNDP (http://europeandcis.undp.org/home/show/707ADE83-F203-1EE9-BF3D08FC372EA619, accessed 9 November 2011).

UNECE (in press). No one left behind. Good practices to ensure equitable access to water and sanitation in the pan-European region. Geneva, UNECE.

UNICEF (2011). Water, sanitation and hygiene [web site]. (http://www.unicef.org/wash/index_3951.html, accessed 25 May 2011).

United Nations (2010). Millennium Development Goals Report 2010. New York, United Nations Department of Economic and Social Affairs.

United Nations Human Rights Council (2010). Human Rights Council resolution A/HRC/15/L.14 on human rights and access to safe drinking water and sanitation. Geneva, United Nations Human Rights Council (http://ap.ohchr.org/documents/dpage_e.aspx?si=A/HRC/15/L.14, accessed 21 January 2012).

United Nations Millennium Project (2005). Investing in development: a practical plan to achieve the Millennium Development Goals. New York, UNDP.

Water Aid (2010). Abandoning open defecation: comparison and adaptation of social change dynamics. Accra, Water Aid Ghana (http://www.wateraid.org/documents/plugin_documents/social_transformation_study_briefing_note.pdf, accessed 7 November 2011).

Water Supply and Sanitation Collaborative Council (2010). Global Sanitation Fund: principles [web site]. Geneva, Water Supply and Sanitation Collaborative Council (http://www.wsscc.org/global-sanitation-fund/principles, accessed 6 July 2010).

WHO (2007). Housing, energy and thermal comfort: a review of 10 countries within the WHO European Region, 2007. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0008/97091/E89887.pdf, accessed 7 November 2011).

WHO (2009). WHO Guidelines for indoor air quality: dampness and mould. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0017/43325/E92645.pdf, accessed 7 November 2011).

WHO (2010a). News release: access to safe drinking water improving; sanitation needs greater efforts. Geneva, World Health Organization (http://www.who.int/mediacentre/news/releases/2010/water_20100315/en/index.html, accessed 25 May 2011).

WHO (2010b). Technical and policy recommendations to reduce health risks due to dampness and mould. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0015/121425/E92998.pdf, accessed 7 November 2011).

WHO (2011a). Towards environmental health inequality reporting. Report of an expert group meeting, Bonn, Germany, 25–26 October 2010. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0013/130243/e94628.pdf, accessed 7 November 2011).

WHO (2011b). Small-scale water supplies in the pan-European Region: background, challenges, improvements. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0018/140355/e94968.pdf, accessed 7 November 2011).

WHO (2011c). Water sanitation and health: water supply, sanitation and hygiene development. Geneva, World Health Organization (http://www.who.int/water_sanitation_health/hygiene/en/, accessed 25 May 2011).

WHO/UNICEF JMP for Water Supply and Sanitation (2010). Progress on sanitation and drinking-water: 2010 update. Geneva, World Health Organization (http://www.unwater.org/downloads/JMP_report_2010.pdf, accessed 7 November 2011).

WHO/UNICEF JMP for Water Supply and Sanitation [web site] (2011). Geneva, World Health Organization; New York, UNICEF (http://www.wssinfo.org/about-the-jmp/introduction/, accessed 7 November 2011).

Zacharasiewicz A et al. (2000). Symptoms suggestive of atopic rhinitis in children aged 6–9 years and the indoor environment. Allergy, 55(10):945–950.

3

INJURY-RELATED INEQUALITIES

CHAPTER

Chapter 3. Injury-related inequalities54

CHAPTEr 3. INjury-rElATEd INEquAlITIESIngrid Fast, Matthias Braubach, Francesco Mitis, Lucie Laflamme

Key messagesIn general terms, unintentional injuries related to or influenced by environmental conditions are a major challenge for public health.• Unintentional injuries are the third leading cause of death in Europe and the associated

morbidity puts a high level of demand on health systems. • Unintentional injuries are one of the causes of mortality and morbidity with the steepest

socioeconomic gradient, affecting lower-income people and areas to a greater extent.• Unintentional injuries can be prevented through a range of evidence-based measures.

This chapter specifically reports on sex- and age-related inequalities, as well as on differences by national income levels, for four injury-related inequality indicators, highlighting the following factors.• Males sustain far more fatal road traffic injuries (RTIs), falls, and poisonings than females, and

the same applies to non-fatal work-related injuries. • The strongest sex-related inequality is found for RTIs, where incidence among males is four

times greater than among females.• Age-related inequalities exist but are not systematic; they differ in magnitude and in direction

across indicators. • The highest age-related disparities are found for fatal falls, where rates among the elderly are

extremely high. By contrast, work-related injury rates do not vary remarkably with age in many countries.

• At country level, low- and middle-income countries have generally higher RTI and poisoning mortality rates than high-income countries.

• Country variations in injury-related inequalities can be very significant, with differences between lowest and highest inequality levels often fivefold or more. However, these differences strongly depend on the respective inequality indicator.


INTrOduCTIONInjuries are a leading cause of mortality worldwide, responsible for 9% of all deaths globally and 7% in the WHO European Region (WHO, 2011). They are also responsible for a great deal of morbidity; injured people are frequently admitted to hospital and, even more frequently, receive care through outpatient visits and medical consultations and treatments (see Fig. 20). Although their consequences are often minor, some injuries may have long-lasting and permanent negative consequences – not only physical (such as life-long disability) but also psychological (such as post-traumatic stress disorder (PTSD)) or social. Superficial or open wounds, contusions, dislocations and fractures are among the most common types of injuries sustained (Sethi et al., 2006a).

Fig. 20. Pyramid of unintentional injuries

1 fatality

36 hospital admissions

185 hospital outpatients

98 other medical treatments

Source: data from Bauer and Steiner, 2009, adapted.

Although injury rates have fallen slightly in Europe during the past decade, there is broad scope for improvement: both global injury rates and the unequal distribution of injuries within and between countries can be tackled and reduced (Laflamme, Burrows and Hasselberg, 2009; Laflamme et al., 2009). There is accumulated and convincing evidence that unintentional injuries are highly preventable (WHO, 2007a), and there is a growing body of knowledge suggesting that inequalities in injury rates can be prevented and reversed (Laflamme, Burrows and Hasselberg, 2009; Laflamme et al., 2009).

This is of the utmost importance, since injuries have become one of the causes of mortality with the steepest social gradient. Map 2 below shows that this is also reflected in the stark mortality rate differences between countries.

There is also increasing evidence from several European countries that inequality-related differences in injury morbidity can be just as stark. These inequalities arise for different reasons, including differences in exposure, in access to means of prevention (and protection) and in injury consequences (Laflamme, Hasselberg and Burrows, 2010). Inequalities have been observed between different sociodemographic groups – based on individual age, sex, and SES – as well as between locations such as country, region and area of residence. Such wide disparities in injuries have been highlighted in the WHO European Region when comparing low- and middle-income countries to high-income ones (Sethi, Mitis and Racioppi, 2010). If all countries’ data were at the level of the lowest national injury-related mortality rate, 508 313 lives could be saved each year within the WHO European Region (accounting for 68% of all injury-related death cases) (Sethi et al., 2006a).

To better contextualize and highlight the inequalities in injuries occurring in Europe, this chapter presents data on inequalities related to the incidence of non-fatal work-related injuries, fatal RTIs, fatal poisonings and fatal falls, stratified by age and sex, and – for RTIs, poisonings and falls – by national income levels. Additional information is also provided concerning the state of knowledge of sociodemographic parameters that – unlike other inequality indicators covered in this report – could not be addressed with the available data.


map 2. Standardized mortality rates for unintentional injuries and poisonings/100 000 population (last year of reporting)

<30

<50

>=50

<20

no data

European Region average 44.64

Mortality rate /100 000

Source: data from European Health for All database (variable 7060), 2011

dATA ANd mETHOdS The data used in this chapter were extracted in spring 2011 from a range of international databases provided by Eurostat and WHO (Eurostat, 2011; European detailed mortality database (DMDB), 2011; European mortality database (MDB), 2011; European Health for All database, 2011). In addition, data from the updated global burden of disease (GBD) (WHO, 2011) were used to supplement the RTI, poisoning and fall information, enabling a closer look at inequalities between high-income and low- and middle-income countries, stratified by sex and age group. No further information on income-related inequalities was available.

Throughout the chapter, sex ratios (SRs) are used to represent the relative difference in incidence or mortality rates between males and females; for example, a value of 3:1 indicates that incidence among males is three times higher than among females.

For the inequality indicators on RTIs, poisonings and falls, it proved very difficult – and sometimes impossible – to generate representative subregional results because of a lack of adequate data. Therefore, the subregional data reported often represent the average of the national rates of the countries covered by the respective region, and this restriction is clearly marked in each affected figure. Subregional data that give accurately representative aggregated results (often provided by Eurostat databases, as in the case of work-related injuries) do not include this indication.

work-related injuriesEurostat provided all data on work-related injuries, defined as “accidents at work resulting in more than 3 days’ absence from work” (Eurostat, 2011). An accident at work is defined as “a discrete occurrence in the course of work which leads to physical or mental harm”. This definition includes work-related injuries that occur both inside and outside the work premises (even if caused by a third party or during transport) and cases of acute poisoning. Injuries caused by accidents on the way to or from work (commuting accidents), occurrences with a purely medical origin (such as a heart attack at work), occupational diseases and general work-associated conditions such as overexertion and musculoskeletal effects were excluded.


Work-related injuries are the only indicator based on non-fatal injury data. Injury incidence rates for males and females (Eurostat code hsw_aw_inasx) were calculated, along with the development of injury incidence for males and females from 1998 to 2006 (Eurostat code tsiem 090). In addition, injury incidence was determined for three age groups: younger (up to 24 years), middle (25–54 years) and older (55 years and above) (based on Eurostat code hsw_aw_inaag). To extend the evidence base, data on occupational death rates from the European Commission and the European Health for All database were also included.

Fatal rTIsData from the DMDB on the number of RTI-related deaths/100 000 population were used to display inequalities in RTI rates by four age groups and to assess differences by sex. The database, further information and definitions are available online on the WHO Regional Office for Europe web site (http://data.euro.who.int/dmdb/) and cover up to 47 of the WHO European Member States.

Data from the DMDB – WHO International Statistical Classification of Diseases and Related Health Problems, tenth revision (ICD-10) codes V01–V79 and V82–V89 (WHO, 2010a) – were compiled to reflect cases of road traffic-related death, excluding mortality related to train or air transport but including cases of death among cyclists and pedestrians. However, if the necessary information was not available, DMDB data on overall transport mortality were used, based on mortality tabulation list 1 (MTL1) of ICD-10 and integrating data from countries still reporting through the ICD-9 system (MTL1 code 1096).

For an assessment of intracountry variations of transport-related mortality, data from the MDB on deaths from transport accidents (code 7120) – available for few European Member States – were used.

Fatal poisoningData from the DMDB on the number of fatal poisonings/100 000 population were used to show inequalities by sex and the related ratios for all poisonings, as well as for alcohol-related poisonings specifically. Age group-specific poisoning death rates were calculated for the most relevant poisonings.

Data were taken from the DMDB (ICD-10 codes X40–X49 and MTL1 code 1100). The database, further information and definitions are available online on the WHO Regional Office for Europe web site (http://data.euro.who.int/dmdb/) and cover up to 46 of the WHO European Member States.

For an assessment of intracountry variations of poisoning mortality, MDB data on the standardized death rates from accidental poisoning (code 7420) – available for few European Member States – were used.

Fatal fallsData from the DMDB on the number of fatal falls/100 000 population were used to calculate inequalities by sex and the related ratios for the full population, as well as for age group categories. Age-specific data show the trend of fall mortality over the life course.

The data were taken from the DMDB (MTL1 code 1097). The database, further information and definitions are available online on the WHO Regional Office for Europe web site (http://data.euro.who.int/hfamdb/) and cover up to 46 of the WHO European Member States.


rESTrICTIONS ANd dATA lImITATIONS There were four major restrictions on the treatment of the data.

lack of data and data coverageAge- and sex-stratified data on work-related injuries and mortality could only be obtained from Eurostat, which has data from a limited number of WHO European Member States. In the data provided by Eurostat on work-related injuries, stratification by sex was only available at country level for 17 EU countries.

For some countries, the latest reported data are several years old and current injury rates are unknown. This hampered an accurate comparison of trends over time and made comparative assessments difficult.

quality of dataData from the MDB and DMDB must be used with caution for intercountry comparisons because for some countries mortality rates may be biased due to the underreporting of death cases, particularly in the central Asian republics, the Caucasus countries and some countries in the Balkan region (MDB, 2011; DMDB, 2011).

lack of country data on individual socioeconomic determinantsCountry-specific information on injuries stratified by SES variables (income, education, and so on) or other relevant dimensions such as ethnicity is not available from the databases used, and would have provided additional information on inequalities in injury rates. Wherever possible, existing empirical evidence and country-level income data were used to cover the socioeconomic dimension of injury-related inequality.

Outcome heterogeneityThe four injury-related inequality indicators studied are heterogeneous groups. Within each of them there are different health outcomes caused by a variety of mechanisms; for example, different categories of road users are associated with different levels of injury risk, and some population groups are particularly affected by alcohol poisoning rather than by other types of poisoning. Variations and differences related to the injury types can be large, implying that there might be underestimations of differences between groups, and that some existing inequalities are possibly not reflected by the data.

INEquAlITIES IN wOrK-rElATEd INjurIES

IntroductionIn Europe alone there are as many as 300 000 work-related deaths every year, and work-related injuries have been defined as the biggest health risk in the working environment, before exposure to noise and carcinogens (WHO, 2002). There is ample evidence from the scientific literature – European and international – that there are substantial inequalities in work-related injury rates, varying especially but not exclusively by occupation. Since occupation and SES are strongly correlated, differences in work-related injury rates can also be found based on individual SES measures such as education and income level (WHO, 2010b; Siegrist et al., 2010; Costa and D’errico, 2006).

Another source of inequality in work-related injury rates is sex: males have higher injury rates than females. This can be explained by differences both in type and amount of occupational exposure and in individual behaviour and vulnerability (WHO, 2010b). Even individual age can come into play as a


source of inequality, with younger and older workers identified as risk groups for their expected lack of experience in the former case (European Agency for Safety and Health at Work, 2011) or their age-related impairment in the latter (European Commission, 2009). But reviews indicate that it is in rather specific circumstances and for specific types of injuries that age can be expected to play a role (Laflamme and Menckel, 1995). Age can, in fact, be not only an aggravating but also a preventive factor.

Work-related injuries are highly preventable through modification of the work environment itself, of prevailing working conditions and also, to some extent, of the working climate and work culture (Council Directive, 1989; WHO, 2010b). Because occupational exposure varies greatly across occupational groups and sectors, some occupations are at far greater risk than others and so are some sectors of activity.

Indicator analysis: inequalities by sex, age and occupationUsing country-based Eurostat data, it can be observed that males have higher work-related injury incidence rates than females in all countries, with an average male–female SR of 2.3:1 for the EU15. The highest ratios are 3.6:1 and 3.5:1 in Luxembourg and Greece, and the lowest is found in Sweden at 1.7:1. For males, the incidence rate of injuries/100 000 ranges from 1145 (Sweden) to 5935 (Spain), and for females from 605 (Greece) to 2446 (Spain) (see Fig. 21).

Fig. 21. work-related injury incidence rate/100 000 population in employment by sex (2007)

0

1000

2000

3000

4000

5000

6000

7000

8000

Sw

eden

Gre

at B

ritai

n

Irela

nd

Gre

ece

[a]

Aus

tria

Italy

Den

mar

k

Net

herla

nds

[a]

Nor

way

Finl

and

Bel

gium

Ger

man

y

Sw

itzer

land

Luxe

mbo

urg

Fran

ce

Por

tuga

l

Spa

in

EU

15 [b

]

Inci

denc

e ra

te /1

0000

0

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Rat

io o

f inc

iden

ce ra

tes

(mal

e:fe

mal

e)

MaleFemaleSex ratio

Source: data from Eurostat, 2011.Notes: [a] data for Greece and Netherlands from 2006; [b] EU15 data based on value for Great Britain replacing unknown value for United Kingdom.

Sex-related differences are also reproduced at economic sector level (see Table 4). Risk of injury to males outweighs that to females by the highest proportion in the construction sector (SR 3.6:1); this is also the sector with the highest injury incidence globally. The SR is lower in sectors such as transport and communication (1.4:1) and hotel and catering (1.3:1). A national example of work-related injury SR differences by sector is provided for Croatia in Annex 1.

Fig. 22 shows the development of the incidence of work-related injuries from 1998 to 2006, the last year of reporting to Eurostat. The bars show the incidence changes for males and females, based on an index value of 100 for the baseline year of 1998. In most EU countries, the occurrence of work-related injuries decreased strongly for both sexes, although with a sharper decline for males than for females. In 2006, the work-related injury index for females was at 83 for EU27, whereas for males the index dropped to 77. This corresponds to injury rate reductions by 17% and 23% since 1998, respectively. In a few countries the rate actually increased for one sex (Ireland, Cyprus, France and Lithuania) and thus indicates an increasing sex-related inequality during the time period covered, while in Estonia, the


rate strongly increased for both sexes. However, there are various countries where, although rates are declining for males and females, there is a strong sex-related difference in the relative reduction (such as Denmark, Czech Republic and Poland).

Table 4. Standardized incidence rate of accidents at work by economic sector and sex (average for 2005–2007)

Economic sector Incidence/100 000 population in employment ratio of incidence rates

male Female (male:female)

Construction 6172 1723 3�6:1Electricity, gas & water supply 2046 715 2�9:1Manufacturing 4003 1855 2�2:1Financial intermediation; real estate 1972 934 2�1:1Wholesale and retail trade; repair of motor vehicles, motorcycles and personal and household goods

2811 1440 2:1

Agriculture, hunting and forestry 4847 2963 1�6:1Transport, storage and communication 2882 2013 1�4:1Hotel and catering 3238 2585 1�3:1

Source: data from Eurostat, 2011.

Fig. 22. work-related injury incidence rate/100 000 population in employment by sex (2006) in index points

Note: smaller font size chosen (8 instead of 9)Country names may have to be updated in lay out version

EU27 countries

NMS12 countriesEstonia

RomaniaLithuania

MaltaPoland

CyprusSloveniaCzech RepublicHungaryBulgaria

Slovakia

EU15Ireland

Netherlands [a]SpainFinlandSwedenLuxembourgDenmark

FranceUnited KingdomPortugalAustriaItalyGermanyBelgiumGreece

40 50 60 70 80 90 100 110 120 130 140Index points (1998=100)

FemaleMale

Source: data from Eurostat, 2011.Note: [a] Netherlands: value = 100 (2005).


Fig. 23 illustrates age-related inequalities in work-related injury rates. In many instances, at country level, workers from the highest age group experience the lowest incidence of work-related injuries (EU15 average: 1917/100 000). Noteworthy exceptions are the Netherlands (4533/100 000), Luxembourg (4897/100 000) and Norway (2947/100 000) where incidences among the highest age group exceed those among younger workers.

Fig. 23. work-related injury incidence rate/100 000 population in employment by age group (2007)

0

1000

2000

3000

4000

5000

6000

7000

8000S

wed

en

Irela

nd

Gre

at B

ritai

n

Net

herla

nds

[a]

Gre

ece

[a]

Den

mar

k

Finl

and

Bel

gium

Aus

tria

Ger

man

y

Italy

Fran

ce

Por

tuga

l

Luxe

mbo

urg

Spa

in

EU

15 [b

]

Nor

way

Sw

itzer

land

Inci

denc

e ra

te/1

0000

0

<18–24 years25–54 years55 years +

Source: data from Eurostat, 2011.Notes: [a] data from 2006; [b] EU15 data based on value for Great Britain replacing unknown value for United Kingdom.

Workers in the middle age group rank second (average EU15 incidence: 2743/100 000); the highest incidence rate is found in Portugal (4539/100 000) and the lowest in Great Britain (914/100 000). Workers in the youngest age group for their part have the highest incidence rates of work-related injuries (average EU15 incidence: 3172/100 000), with the highest rate in Spain (7923/100 000) and the lowest in Sweden (686/100 000). Compared to sex-related inequalities, age-related inequalities in work-related injury rates are not very pronounced in general, but for individual countries (such as France, Italy, Luxembourg, Portugal and Spain) the incidence differences are remarkably high.

However, looking at fatal work-related injuries within the EU, the age-related inequality pattern changes, and the sex-related inequality pattern is enhanced. Age-stratified mortality data for the year 2000 indicate that fatal work-related injuries are far more likely to affect older workers than their younger counterparts (Eurostat, 2011). Fatal injury data stratified by sex show that in sectors with the highest mortality rates (agriculture, manufacturing, construction and transport), 95% of all fatal injuries occurred among males (European Commission, 2009).

The data presented in this chapter are limited to EU countries and it remains unclear whether the patterns observed herein apply to the same extent to other countries of the WHO European Region. However, the European Health for All database (parameter 4070) presents general data on work-related mortality rates for 50 countries which range from 0.2/100 000 to 10.7/100 000. The data indicate strong regional differences in work-related mortality rates (Euro 1 countries showing the lowest mortality rates while the highest are found in Euro 3 countries), but cannot be stratified by age or sex.

Target groups for actionAmong the countries contributing data, males sustain more work-related injuries than females and thus represent a target group for action. This is true for data within each sector and aggregated from all economic sectors, but is mostly relevant in manual labour-intensive sectors (see Table 4). However, without data on occupational exposure, it is not possible to determine the specific occupations at higher risk or indeed the sources of the hazards.


It is important to note that some health problems that are more common among females are missing from the material under consideration: this is because the definition of work-related injuries includes only those occurring during the course of work. Within the health sector, for instance, where females make up a greater proportion of employees than males, musculoskeletal problems are highly prevalent among nurses and aides but are not considered to be direct work-related injuries.

A closer look at data stratified by occupation and economic sector would also help to broaden understanding of age-related inequalities, which are less clearly expressed when looking at the total number of work-related injuries.

Although inequality dimensions other than age and sex could not be considered in the analyses conducted, it is important to note that workers with low SES and migrant workers are known vulnerable groups that suffer traumatic injuries more often, a fact that has been attributed to their generally worse health status. But poorer working conditions and work environments may also contribute to their higher risk of work-related injury (European Agency for Safety and Health at Work, 2011).

Health implications Work-related injuries most often result in sick leave, and there are more injuries that result in sick leave among older than younger workers. Longer durations of sick leave in general are more prevalent among males and the elderly working population (European Commission, 2009). However, work-related injuries can also lead to long-lasting disabilities and, as mentioned above, even have fatal outcomes. Disability is a threat for workers of all age groups: among adults, around 6% of disabilities are a consequence of work-related injuries (Tuchsen et al., 2009).

Regarding the types of injuries sustained, from 1997 to 2005 a decrease has been observed in the incidence of wounds and superficial injuries as burns, scalds and frostbite. By contrast, occurrences of dislocations, sprains and strains are on the increase, as well as multiple and internal injuries resulting in death (European Commission, 2009).

Work-related injuries also affect a worker’s social life and future prospects; for instance, disabilities that reduce work performance or capacity to remain in an occupation lead not only to loss of employment but also to an impoverished quality of life. The long-term absence from work or long-term unemployment of a breadwinner might even put an entire family in a precarious situation.

Conclusions and suggestions Although the available data are representative of only a limited number of countries, some trends and patterns are noteworthy. One is that there are clear sex-related inequalities in the incidence of work-related injuries, to the disadvantage of male workers. However, in recent years, decreases in work-related injuries have been more remarkable among male workers than female workers.

Also striking is the heightened risk of occupational injuries among younger workers, which is strongly expressed in several countries. By contrast, older workers are over-represented when analysing occurrences of fatal work-related injuries.

The risk of work-related injury can be significantly reduced by the design and layout of safer working environments, methods, and instruments. Work to build up and promote these factors is likely not only to reduce the probability of injury but also to narrow down the level of inequality between males and females employed in similar occupations (Eurostat, 2011). However, more reliable reporting of work-related injuries in non-EU countries would extend the available information on the incidence of work-related injuries and their distribution among different segments of the workforce within the WHO European Region, as well as helping to identify target groups for action.


Suggested mitigation actions are:• implementation of national programmes on occupational safety and health as suggested by

ILO Conventions 155 (ILO, 1981) and 161 (ILO, 1985) and the Workers’ health: global plan of action (WHO, 2007b), aiming at universal coverage of prevention-oriented occupational health services for all workers;

• further implementation and enforcement of the Community strategy 2007–2012 on health and safety at work (European Commission, 2007) and similar relevant policies applying outside the EU;

• thorough induction training of younger and new workers on workplace hazards and respective safety measures;

• wherever possible, adapting job demands and conditions to the physical and mental capacity of workers;

• provision of basic occupational health services to workers from vulnerable groups (younger, elderly, disabled, migrant, pregnant, and so on) and those in high-risk sectors (such as small enterprises, construction, manufacturing, health care, agriculture, mining and informal);

• better reporting on work-related injuries through occupational health surveillance programmes, including their distribution by age, sex and occupation.

INEquAlITIES IN FATAl rOAd TrAFFIC INjurIES (rTIS)

Introduction An increasingly mobile society with growing traffic volume bears a rising risk of RTIs, including injuries related to both motorized and non-motorized means of transport. As the leading cause of deaths among adolescents and young adults, and an increasingly important cause of injury and death among children, RTIs have among the most serious consequences of all injuries (Sethi et al., 2006b). In 2010 alone RTIs resulted in 107 850 deaths in the WHO European Region (WHO, 2009). In western Europe an estimated 25% (12–59%) of RTIs are caused by environmental factors such as urban structure and road design (Pruss-Üstun and Corvalán, 2006). Alongside environmental risks, alcohol consumption and speeding rank highly as factors associated with both crash occurrence and crash severity. It has been estimated that around one in four injuries is caused by drink–driving (Directorate General for Health and Consumer Protection, 2006).

RTIs are not randomly distributed among areas and groups in society (Laflamme and Diderichsen, 2000; Laflamme, Hasselberg and Burrows, 2010). In Europe, residents of the Euro 3 countries are more at risk than those of other regions: as a result of these states’ late transition to market economies and a resulting increase in motorization without the matching infrastructure, road safety tends to be low (Racioppi et al., 2004). Alongside well-known differences between, for example, urban and rural areas, there are also remarkable inequalities between socioeconomic groups, to the detriment of disadvantaged ones (WHO, 2009; Laflamme, Hasselberg and Burrows, 2010). Within countries, a high proportion of the RTI-related burden of injury is carried by those with low SES, in part because they may not have equal access to safe transport options (Kay et al., 2011).

In addition, males are at far greater risk of RTI than females and so are people in the younger and middle age categories. These inequalities may, however, vary depending on the category of road user (Laflamme, Hasselberg and Burrows, 2010).


Indicator analysis: inequalities by country income, age and sexBased on recent GBD data (WHO, 2011), Fig. 24 shows the inequalities of RTI mortality rates between high-income countries and low- and middle-income countries in the WHO European Region. Among all age groups and both sexes there is much higher RTI-related mortality in low- and middle-income countries. The relative inequality between countries with high national income levels versus those with low and middle national income levels is lowest in the 15–29 year age group in both males (low- and middle-income country to high-income country ratio of 1.6:1) and females (ratio of 1.8:1). The highest inequalities are found for children, with country income ratios of 3.4:1 (males, 0–4 years) and 3.3:1 (females, 5–14 years). Despite the much higher mortality rates in males, the ratio of high-income to low- and middle-income countries is on a similar level for males and females, showing that both sexes are, in relative terms, similarly disadvantaged by national income levels.

Fig. 24. rTI mortality rate/100 000 population by national income level, sex and age in the wHO European region (2008)

0

1000

2000

3000

4000

5000

6000

7000

8000

0–4 5–14 15–29 30–44 45–59 60–69 70–79 80+Age group

Mor

talit

y ra

te/1

0000

0

Males, high-incomecountries

Males, low/middle-incomecountries

Females, high-incomecountries

Females, low/middle-income countries

0

0.5

1

1.52

2.5

3

3.5

4

0–4 5–14 15–29 30–44 45–59 60–69 70–79 80+Rat

io o

f mor

talit

y ra

tes

(low

/mid

dle-

inco

me:

high

-inco

me

coun

tries

)

National income level ratio – males

National income level ratio – females

Source: calculated from GBD 2008 data (WHO, 2011).

A compilation of standardized mortality rates/100 000 by age group (0–14, 15–24, 25–64, 65+ years) is shown in Fig. 25a and the same calculation is shown in Fig. 25b using data on all transport-related mortality rates for those areas where RTI-specific data were not available.7

7 RTI data used for Fig� 25a are based on the DMDB (ICD-10 codes V01–V79 and V82–V89) and refer to all fatal injuries occurring on the roads, including those to pedestrians and cyclists� Fatal injuries to passengers of trains, aircraft and ships are excluded� Fig� 25b uses data on “Transport mortality” from the DMDB (MTL1 code 1096) for countries where detailed data on RTIs were not available�


Fig. 25a. Age-standardized mortality rate/100 000 population from rTIs by age group (last year of reporting)

Source: DMDB, 2011

we need to see how this works in the layed out report

0 5 10 15 20 25 30 35

All countries [a](n=37)

Euro 4 countries [a]Croatia

MontenegroSerbiaIsrael

Euro 3 countries [a]Republic of Moldova

KyrgyzstanUzbekistan

GeorgiaAzerbaijan

Euro 2 [a]PolandCyprus

LithuaniaSloveniaRomaniaBulgariaEstonia

Czech RepublicSlovakia

LatviaHungary

Malta

Euro 1 countries [a]Belgium

ItalyDenmark

FranceAustria

GermanyLuxembourg

PortugalSpain

NorwayFinlandIreland

United KingdomNetherlands

SwedenIceland

Mortality rate/100 000

0–14 years

15–24 years25–64 years

65 years +

Source: data from DMDB, 2011.Note: [a] average of national rates.


The aggregated data for all countries show that the lowest RTI mortality rate is among children, with an average of 1.7 deaths/100 000, while the mortality rates for all other age groups are much higher (see Fig. 25a). In 20 out of 37 countries the highest mortality rate falls in the 15–24 year age group (total average of 11.7/100 000), which is also the highest total average rate, followed by the highest age group (65 and older), at 11.5/100 000. Those aged over 65 years have the highest RTI mortality rates in 14 countries.

Fig. 25b. Age-standardized mortality rate/100 000 population from all transport injuries by age group (last year of reporting)

we need to see how this works in the layed out report

0 5 10 15 20 25 30 35

All countries [a](n=10)

Euro 4 countries [a]

Albania

MKD [b]


Russian Federation

Kazakhstan

Belarus

Ukraine

Armenia


San Marino

Greece

Switzerland

Mortality rate/100000

0–14 years15–24 years25–64 years65 years +

Source: data from DMDB, 2011.Notes: [a] average of national rates; [b] MKD: ISO code for the former Yugoslav Republic of Macedonia.

Fig. 25b indicates a roughly similar situation for all transport mortality. All age groups have somewhat higher mortality rates than in the RTI-specific data, but – as with the RTI mortality data – child mortality rates are much lower (at an average of 2.9 deaths/100 000) than the rates for other age groups. The highest average mortality rate is found for the 15–24 year age group (18.6/100 000), but it is only a little lower for the 25–64 year age group (15.6/100 000) and the oldest age group (14.8/100 000).

Although the average aggregated data for all countries show no large differences between the three oldest age groups, Figs. 25a and 25b indicate that age variations can be rather different within countries. In fact, the relatively low mortality rate among children aged up to 14 years is the only commonality for all countries. In addition, countries show strong variations in the national level of transport-related mortality, with Albania, Kazakhstan, Kyrgyzstan, the Russian Federation and San Marino showing very high rates, exceeding 25/100 000 for at least one age group.

More detailed data for some countries show that intracountry variations in RTIs exist and can be quite diverse. Table 5 presents selected examples of countries with wide regional variations in RTI mortality (Spain: factor 3.7, Russian Federation: factor 3.8) and countries with lower regional disparities (Finland: factor 2.1, Hungary: factor 1.3).

Fig. 26 presents the differences by sex in road traffic-related mortality, showing the SR of fatal RTIs for different age groups. The mortality from RTIs is higher in males throughout all age groups, indicated by median values higher than 1.


Table 5. Intracountry differences in transport-related standardized death rates/100 000 population for selected countries (last year of reporting)

Finland (2009) Hungary (2009) russian Federation (2001)

Spain (2008)

Minimum mortality rate 3�7 8�2 13�0 2�6Average mortality rate 6�1 9�7 25�8 7�1Maximum mortality rate 7�9 10�7 49�1 9�5Factor between minimum and maximum rate

2�1 1�3 3�8 3�7

Source: data from MDB, 2011.

The inequality in RTI mortality levels between males and females is greatest for young adults, reaching a more than fivefold difference in those aged 25–29 years. After the age of 29 the inequality between the sexes decreases moderately to a SR between 2:1 and 3:1 in the older age groups. Analysis of the overall transport injury data from 47 countries shows that the average transport-related mortality is 15.7/100 000 for males and 4.2/100 000 for females, indicating that the male mortality rate is almost four times higher (DMDB, 2011).

Fig. 26. Srs by age group for rTI mortality (last three reporting years)

Age group

0

2

4

6

8

10

12

Rat

io o

f mor

talit

y ra

tes

(mal

e:fe

mal

e)

0-14 15-19 20-24 25-29 30-44 45-59 60-64 65-69 70-74 75+

20

32

Source: data from DMDB, 2011.Note: country coverage as for Fig. 25a.

Reversed sex-related inequalities with higher mortality rates in females can be observed in only three countries, but the difference is much smaller. Such inequality is found in the youngest age groups (0–14 years and 15–19 years) in Azerbaijan and Denmark; in the older age groups higher mortality rates in females are reported by Azerbaijan and Montenegro.

Target groups for action Personal characteristics which influence vulnerability to RTIs are sex (males have a far higher incidence rate than females) and age (young adults and older road users have the highest incidence rates). Young adults also have higher rates of mortality from RTIs as car drivers. The age- and gender-associated lifestyle factors for this group, such as speeding or overconfidence in males, or driving under the influence of alcohol and drugs, raise the incidence of RTIs (Peden et al., 2004; European Commission, 2011).


In old age (65+ years), incidence of RTIs is determined by functional limitations (increased crash risk) and physical vulnerability (increased injury severity), which result in a higher risk of fatality than injury (European Commission, 2011).

Fig. 27 shows the distribution of RTI mortality by road user category, indicating that about one-half of all RTI mortality in the WHO European Region is suffered by car occupants. However, pedestrians account for 28% of all deaths from road traffic. A national example of transport mortality for different road users stratified by sex and age is provided for Malta in Annex 1.

Fig. 27. distribution of rTI fatalities by road user category for the wHO European region

Occupants of four-wheeled vehicles,

53%

Riders of motorized two- or

three-wheeledvehicles, 8%

Cyclists, 3%

Pedestrians, 28%

Other, 2% Unspecified, 5%

Source: data from WHO, 2009, adapted. Data have been reported by countries for different years.

In urban areas, pedestrians are the most vulnerable road users, because they are usually not or scarcely protected. Unprotected road users are vulnerable to the greater mass and speed of motorized vehicles (European Commission, 2011). Among pedestrians, people aged below 10 and over 65 years are at especially high risk. Pedestrians have the highest fatality rate in urban areas and account for 33% of fatal RTIs. Pedestrians with low SES are at especially high risk because they have least access to safer modes of transport (Kay et al., 2011). In rural areas, however, car drivers have the highest incidence of fatal accidents (43%), while pedestrians account for only 9% of fatal RTIs in rural settings (European Commission, 2011).

Health implicationsThe health consequences of RTIs vary in severity. The most severe outcomes are fatal injuries, but the fatality rate is relatively low (2% in 2005) in comparison to other outcomes (European Transport and Safety Council, 2007). The recently decreasing fatality rates from RTIs are partially counteracted by an increased number of severe injuries with long-term consequences. More serious health implications can result in long-term or permanent disability such as spinal cord injuries (1.4% of all RTIs), a fractured thighbone (1.3%) and long-term intracranial injuries (1.2%). RTIs often result in short-term disability, such as short-term intracranial injuries (24.6%); open wounds (10.3%), various fractures and internal injuries (<6.3%) are less severe consequences of RTI (Peden et al., 2004). As an overall result of RTIs, 17% of all DALYs are lost (Racioppi et al., 2004).

Psychological effects of RTI include PTSD in all age groups, and the intangible costs and economic strains on affected families are high (Sethi et al., 2006a).


Conclusions and suggestionsThere are inequalities in RTI-related death rates within and between all countries of the WHO European Region. Both age and sex play an important role but affect RTI rates differently: age differences appear to be rather country-specific, while males suffer more road deaths than females in all countries. The higher rate of mortality among males is partly because of greater exposure (they are more likely to be on the roads than females) and partly because of risk-taking behaviour on the roads. Income levels affect the magnitude of RTIs both in relation to country income levels – as indicated by Fig. 24 – and in relation to individual income and SES, as shown in literature on the subject (see Kay et al., 2011; Laflamme et al., 2009; Laflamme, Hasselberg and Burrows, 2010).

In order to tackle inequalities of fatal RTI rates and RTI-related adverse health outcomes, more detailed understanding about the causes of inequalities in the WHO European Region is required. The large national variations of inequality patterns mean that only broad guidelines or recommendations can be made (see, for example, Mock et al., 2004). A good starting point would be to decrease the overall incidence of RTIs, as suggested by the World report on road traffic injury prevention (Peden et al., 2004), the EU’s Policy orientations on road safety 2011–2020 (European Commission, 2010) and the Global plan for the decade of action for road safety 2011–2020 (WHO, 2010c). This includes action on the environmental causes of RTIs, which are determined by, for example, urban structure, road design and land use patterns (Pruss-Üstun and Corvalán, 2006). Changes to the environment such as better urban and street planning, street maintenance, installation of adequate street lighting and general traffic calming have already been shown to be effective interventions (Bunn et al., 2003; Beyer and Ker, 2009).

Nevertheless, targeted and specific approaches are necessary to reduce RTIs in the main risk groups. There is a need to decrease the occurrence of RTIs in young adults – specifically males – and also to reduce the vulnerability of older road users.

Suggested mitigation actions are:• separating motorized vehicles from unprotected and less protected road users such as pedestrians

and cyclists;• implementation of traffic-calming measures to decrease vehicle speed in urban areas;• promoting the use of individual safety measures (such as seatbelt, child car seat and helmet use);• legislation and enforcement of laws that control speed, drink–driving and ensure safety

equipment use;• promotion of safe driving behaviours in general and consideration given to specific campaigns

targeted at main risk groups such as male drivers;• better reporting on RTIs, including their distribution by sociodemographic groups;• implementation of the EU’s Policy orientations on road safety 2011–2020 (European Commission,

2010) and the Global plan for the decade of action for road safety 2011–2020 (WHO, 2010c).

INEquAlITIES IN FATAl POISONING

IntroductionUnintentional poisoning is the third leading cause of death from injury in Europe (Bauer and Steiner, 2009). In 2008, 84 059 people died in the WHO European Region from unintentional poisoning, corresponding to a mortality rate of 9.5/100 000 (WHO, 2011). Poisonings have a wide variety of causes, ranging from chemicals and medicine in the home to alcohol intoxication.


Recent research has identified an association between age and type of poisoning. Among the elderly, the main causes of poisoning are smoke from fire or other gases, and chemicals, pharmaceuticals and food (Centre for Research and Prevention of Injuries, 2011). Among adults in the middle age categories alcohol is the main cause of poisoning. Additional causes lie in a variety of exposures, most – but not all – of which are occupational ones. In young adults, substance abuse (including alcohol) is the main cause of poisoning; this type of poisoning is on the increase. The associated health effects of such poisoning are dependent on the substance (Sethi et al., 2006b). As many as 90% of poisonings among children occur at home: substances involved in childhood poisoning are often pharmaceuticals, cleaning fluids, cosmetics and other chemicals found in and around the home, as well as certain plants which cause poisoning (European Child Safety Alliance, 2011). Carbon monoxide exposure in the home and lead exposure in the general environment are additional sources of poisoning.

Factors associated with higher rates – or consequences – of poisonings are household poverty, living in deprived communities, single parenthood and overall low SES (Sethi et al., 2006a). Among children in particular, socioeconomic inequalities in unintentional poisoning risk can be strong (Laflamme, Hasselberg and Burrows, 2010). It is noteworthy that the risk of injury from poisoning can be particularly high in migrant populations as shown, for instance, in the Netherlands (Stirbu et al., 2006).

Indicator analysis: inequalities by country income, age and sexRecent GBD data (WHO, 2011) indicate that inequalities in mortality rates from poisoning between high-income and low- and middle-income countries are strongly to the disadvantage of the poorer countries (see Fig. 28). The differences are even greater than those found for RTIs (see Fig. 24). Small children (0–4 years) in low- and middle-income countries are especially affected, with a mortality rate more than 20 times higher than those in high-income countries (both males and females).

Fig. 28. Poisoning mortality rate/100 000 population by national income level, sex and age in the wHO European region (2008)

0

2000

4000

6000

8000

10000

12000

14000

0–4 5–14 15–29 30–44 45–59 60–69 70–79 80+

Age group

Mor

talit

y ra

te/1

0000

0

Males, high-incomecountriesMales, low/middle-incomecountriesFemales, high-incomecountriesFemales, low/middle-incomecountries

0

5

10

15

20

25

0–4 5–14 15–29 30–44 45–59 60–69 70–79 80+

Rat

io o

f mor

talit

y ra

tes

(low

/mid

dle-

inco

me:

high

-inco

me

coun

tries

)

National income level ratio – males

National income level ratio – females



Fig. 29. Poisoning mortality rate/100 000 population by sex (last year of reporting)

0

10

20

30

40

50

60

70

Aze

rbai

jan

Arm

enia

Geo

rgia

Uzb

ekis

tan

Rep

ublic

of M

oldo

va

Kyr

gyzs

tan

Ukr

aine

Kaz

akhs

tan

Bel

arus

Rus

sian

Fed

erat

ion

Eur

o 3

coun

tries

[a]

Isra

el [c

]

Mon

tene

gro

[c]

MK

D [d

]

Ser

bia

Cro

atia

Alb

ania

Eur

o 4

coun

tries

[a]

All

coun

tries

[a]

(n=4

6)

Mor

talit

y ra

te/1

0000

0

0

1

2

3

4

5

6

7

Rat

io o

f mor

talit

y ra

tes

(mal

e:fe

mal

e)

Male Female Sex ratio

0

5

10

15

20

25

30

35

Hun

gary

Cyp

rus

[b]

Bul

garia

Mal

ta

Cze

ch R

epub

lic

Slo

veni

a

Slo

vaki

a

Rom

ania

Pol

and

Latv

ia

Lith

uani

a

Est

onia

Eur

o 2

[a]

All

coun

tries

[a]

(n=4

6)

Mor

talit

y ra

te/1

0000

0

0

1

2

3

4

5

6

7

Rat

io o

f mor

talit

y ra

tes

(mal

e:fe

mal

e)

0

5

10

15

20

25

30

35

Aus

tria

Por

tuga

l

Italy

Ger

man

y

Net

herla

nds

Icel

and

Bel

gium

Fran

ce

Spa

in

Sw

itzer

land

Uni

ted

Kin

gdom

Gre

ece

Den

mar

k

Sw

eden

Luxe

mbo

urg

Irela

nd

Nor

way

Finl

and

Eur

o 1

coun

tries

[a]

All

coun

tries

[a]

(n=4

6)

Mor

talit

y ra

te/1

0000

0

0

1

2

3

4

5

6

7

Rat

io o

f mor

talit

y ra

tes

(mal

e:fe

mal

e)

Source: data from DMDB, 2011.Notes: [a] average of national rates; [b] Cyprus: SR=11.0; [c] data reported for fatalities in males only; [d] MKD: ISO code for the former Yugoslav Republic of Macedonia.

Mortality rates for males are much higher, as with the RTI data (Fig. 24), but the mortality rate ratios of low- and middle-income to high-income countries are very similar, indicating that the impact of national income levels is similar for both males and females.

Data from the DMDB reveal that the average mortality rate from poisoning is 9.8 and 2.6/100 000 for males and females respectively – a SR of 3.8:1 to the disadvantage of males. Fig. 29 shows that this increased risk in males is reflected in all countries but expressed at very different levels of magnitude: the SR is lowest in Austria at 1.4:1 and exceeds 6:1 in Cyprus, Greece and Malta. However, the data also


indicate that the intercountry variability of poisoning rates – with the highest occurrences in the Euro 3 subregion – is more remarkable than the general sex-specific variability, which does not differ much between the Euro regions (ranging from an SR of 3:1 in Euro 1 to an SR of 4.1:1 in Euro 2). A national example of poisoning mortality stratified by age, sex and urban versus rural residence is provided for Poland in Annex 1.

Table 6 shows the proportion of alcohol poisoning among all poisonings for those countries reporting such data. Alcohol causes one-third (33.9%) of all poisonings on average, but there are extreme intercountry variations, with the contribution ranging from 3.6% (Belgium) to 78.6% (Poland). The SR of 3.7:1, however, is similar to the SR for all poisonings (3.8:1).

Table 6. Alcohol-related poisoning mortality rate/100 000 population by sex (last year of reporting)

Country Alcohol-related poisoning ratio of mortality rate (male:female)

Percentage of all poisonings related to alcoholTotal male Female

Belgium 0�1 0�1 0�0 5�6:1 3�6

Denmark 0�3 0�4 0�2 2�2:1 7�8

Finland 8�0 13�3 2�7 4�9:1 54�8France 0�3 0�5 0�1 4�5:1 17�3Germany 0�2 0�3 0�1 3�0:1 19�5Ireland 2�1 2�8 1�4 2�0:1 30�4Italy 0�0 0�0 0�0 1�5:1 4�0Luxembourg 3�8 4�8 2�8 1�7:1 56�4Netherlands 0�1 0�1 0�1 0�9:1 7�5Norway 1�0 1�3 0�7 1�9:1 13�2Portugal 0�0 0�0 0�0 2�1:1 11�2Spain 0�1 0�1 0�1 2�6:1 5�9Sweden 1�1 1�5 0�6 2�6:1 28�6United Kingdom 0�4 0�6 0�3 2�3:1 15�1Euro 1 countries [a] 1�2 1�8 0�6 2�9:1 19�7Bulgaria 0�5 0�8 0�3 2�2:1 29�0Czech Republic 1�7 2�6 0�8 3�4:1 56�4Estonia 8�2 13�6 3�7 3�7:1 44�3Latvia 7�0 12�5 2�5 4�9:1 62�6Lithuania 10�0 16�9 4�2 4�1:1 60�5Poland 3�2 5�9 0�8 7�4:1 78�6Romania 1�9 3�1 0�8 3�8:1 42�3Slovakia 2�3 3�9 0�7 5�2:1 74�5Slovenia 0�9 1�7 0�2 6�7:1 28�9Euro 2 countries [a] 4�0 6�8 1�6 4�3:1 53�0Georgia 0�7 1�0 0�4 2�6:1 58�5Kyrgyzstan 8�0 13�3 3�4 3�9:1 62�0Republic of Moldova 5�3 8�3 2�8 3�0:1 43�4Uzbekistan 0�3 0�6 0�1 4�7:1 15�8Euro 3 countries [a] 3�6 5�8 1�7 3�4:1 44�9Croatia 0�5 0�8 0�1 6�5:1 16�7Euro 4 countries [a] 0�5 0�8 0�1 6�5:1 16�7

All countries (n=28) [a] 2�4 4�0 1�1 3�7:1 33�9

Source: data from DMDB, 2011.Note: [a] average of national rates.


An analysis of intracountry regional variation for a subset of 12 countries reporting subnational data shows that the poisoning mortality rates reveal smaller intra- than intercountry differences, with the exception of the Russian Federation where regional rates range from 1.2–130.3/100 000 (MDB, code 7420).

Regarding age-related inequalities, Fig. 30 confirms that there is a specific age profile for each of the various causes of poisoning reported in the DMDB. Generally, rather low mortality rates are found in the youngest age group (0–19 years), followed by increased mortality rates with rising age up to 39 years. The starkest increase is identified for narcotics and psychodysleptics, representing illegal drugs. Thereafter, mortality trends in the different age groups vary by poisoning cause. The elderly are the main risk group for gases and vapours, pesticides and unspecified chemicals, while the rates for narcotics decrease again after the age of 39. Older adults are the main risk group for alcohol, organic solvents and vapours.

Fig. 30. Poisoning mortality rate/100 000 population by age group and cause (last year of reporting)

two figures to appear as one

0

1

2

3

4

5

6

7

Other gases and vapours

Alcohol Narcotics and psychodysleptics[hallucinogens], not elsewhere

classified

Mor

talit

y ra

te/1

0000

0

0–9 years 10–19 years 20–29 years 30–39 years 40–49 years50–59 years 60–69 years 70–79 years 80+ years

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Organic solvents and their vapours

Other and unspeci-fied chemicals

Pesticides Medical drugs excluding narcotics/

psychodysleptics

Mor

talit

y ra

te/1

0000

0

Source: data from DMDB, 2011.

Furthermore, Fig. 30 shows that the total magnitude of age-related inequalities varies by cause of poisoning. The overall highest mortality rates are alcohol-related, characterized by strongly increasing mortality rates with increasing age, to a maximum mortality rate of 5.9/100 000 for the 50–59 year age group. Narcotic substances cause the second highest mortality rates. The rates of fatal poisoning by pesticides are the lowest ones.


Analysis of all poisoning causes by sex (data not shown) indicates that, in general, there is a higher prevalence of poisoning among males than females for almost all substances. The largest differences by sex are seen for alcohol – as shown above – as well as for narcotic substances and non-organic gases and vapours. On the other hand, several countries indicated a higher level of pesticide poisoning among females, including Croatia and Bulgaria (where mortality among females is double that among males) and Portugal (where mortality among females is six times higher).

Target groups for actionMales have higher mortality rates from poisoning than females. People in their fifties (50–59 years) constitute the age group with the highest overall mortality rates, but this is strongly related to alcohol poisoning, which is the most relevant single cause of fatal poisoning. However, seen from a cause-specific perspective, the older population is a key risk group for many poisonings.

By far the highest mortality rates are found in the countries of the Euro 3 region but the Baltic countries – together with Finland – represent another geographical poisoning hotspot.

Low national income levels strongly increase the overall risk of poisoning, but – in contrast to the higher proportions of mortality among adults and the elderly described above – especially affect poisoning rates among small children. However, the DMDB data do not provide information on poisonings by socioeconomic attributes of individual victims, and thus this dimension could not be studied.

Health implicationsThe health effects of poisonings can be wide-ranging in scope and duration because of the variety of poisonous agents. Symptoms of poisoning often include nausea, vomiting, stomach cramps, pain, diarrhoea and bloody stools. Fever, chills, headaches and weakness can also occur, and more severe cases can lead to shock or collapse (Winter Griffith, Moore and Yoder, 2006). Extreme poisoning can lead to death.

Besides acute poisoning related to CO, alcohol or consumption of hazardous items, for example, long-term poisoning such as lead poisoning can occur, which can cause fatigue or sleeping disorders, behaviour changes in adults and children, stomach discomfort, headache, vomiting and weight loss (Winter Griffith, Moore and Yoder, 2006).

Overall, poisoning is among the 15 leading causes of total DALYs lost. 2 156 438 DALYs were lost through poisoning in low- and middle-income countries in the WHO European Region in 2000 (Peden, McGee and Krug, 2002).

Conclusions and suggestionsAlthough injuries from poisoning are preventable to a high degree, they are an important cause of mortality and morbidity and time trends show that they persist. Those most affected by fatal poisonings are males in their fifties, where alcohol plays a major role. Alcohol is also a major concern, as is the use of drugs. However, other sources of poisoning to which people of all ages can be exposed at home, at work or in the general environment, also need close attention regarding their prevention.

For all vulnerable groups, the maintenance or establishment of a suitable network of poison control centres would enable fast and effective help in emergency situations.

Further information about the occurrence of poisonings, underlying causes and vulnerable groups should also be identified and analysed.


Suggested mitigation actions are:• reducing alcohol poisoning by implementation of the Global Strategy to reduce the harmful

use of alcohol (WHO, 2010d);• reducing the risk of child poisoning through preventive measures, including safe packaging

(manufacturers) and safe storing (home safety practice) of dangerous items and substances, as well as initiatives promoting increased parental supervision;

• a ban on production and availability of lethal substances;• ensuring and promoting working and living conditions free from toxic or poisonous materials;• establishment or maintenance of poison control centres and provision of information to the

public on risks of intoxication from the main causes of poisoning;• improving the surveillance systems monitoring the trends, causes and social context of

poisoning cases.

INEquAlITIES IN FATAl FAllS

IntroductionFalls are the second leading cause of unintentional injury deaths globally (WHO, 2010e). In the WHO European Region, 65 991 fall-related deaths occurred in 2008 (WHO, 2011). Falls are also the leading cause of non-fatal unintentional injuries in children (Sethi et al., 2008), although there is a strong association between age and type of fall, and the incidence and the health consequences of falls are by far the highest and most severe among the elderly. There is evidence that non-fatal fall injuries are in general more common among females and fatal ones more common among males (WHO, 2007a). There is also evidence that some types of falls are more representative of people in good physical condition whereas others falls are more typical for people in poorer physical condition, especially among older people (Kelsey et al., 2010). Falls occur in many different manners and circumstances – at work, at home, when commuting, during sports and leisure time, and so on.

Studies on the socioeconomic distribution of fall-related injuries indicate that high SES (measured in terms of education or income, for example) can be associated with a lower rate of falls in different age groups (among children and older people in particular), but the results are not consistent (WHO, 2007a; Sethi et al., 2008). Inequalities in fall-related injuries occur as a result not only of individual differences but also as a consequence of environmental disparities, including the built environment (such as the design and condition of stairs and balconies), and the products people make use of or are exposed to (WHO, 2007a). Unsafe and inadequate housing and living environments lead to higher rates of injury risk.

Indicator analysis: inequalities by country income, age and sexFig. 31 shows income-related inequalities in fatal falls/100 000 between males and females in high-income countries versus low- and middle-income countries. As with the other injury types (see Figs. 24 and 28), low- and middle-income countries have higher rates of fatal falls up to the 60–69 year age group. However, in the two oldest age groups the picture is reversed and the rate of fatal falls in high-income countries exceeds those in the low- and middle-income countries. Income ratios for males and females are very similar throughout all age groups, except for a peak in females in the age group 5–14 years. In all other age groups the rate ratio is slightly higher for males than females.


Fig. 31. Fall mortality rate/100 000 population by national income level, sex and age in the wHO European region (2008)

Source : calculated from GBD 2008 data (WHO, 2011)0

500

1000

1500

2000

2500

3000

3500

0–4 5–14 15–29 30–44 45–59 60–69 70–79 80+Age group

Mor

talit

y ra

te/1

0000

0

Males, high-incomecountriesMales, low/middle-income countriesFemales , high-incomecountriesFemales, low/middle-income countries

0

1

2

3

4

5

6

7

8

0–4 5–14 15–29 30–44 45–59 60–69 70–79 80+

Rat

io o

f mor

talit

y ra

tes

(low

/mid

dle-

inco

me:

high

-inco

me

coun

tries

) National income level ratio – malesNational income level ratio – females


The available data indicate that males have a higher level of mortality from falls than females in all countries of the WHO European Region (see Fig. 32). On average, the fall mortality rate/100 000 is 8.8 in males and 3.7 in females (SR 2.3:1) but as shown, intercountry differences are stark; the highest differences by sex are identified in the Euro 3 countries with an average SR of 4.7:1, along with the highest national SRs of 6:1 and higher (in Azerbaijan and the Republic of Moldova). High levels of sex-related inequality are also found for some Euro 2 countries such as Bulgaria and Romania, which show SRs of more than 5:1, while lower levels are identified for Euro 1 countries, with the lowest disparities in Iceland and Luxembourg (SR 1.2:1). In consequence, the SRs of fatal falls show a wide range from 1.2:1 to 7.3:1.

For both males and females, age-based differences reveal relatively low fall mortality rates – below 3/100 000 for both sexes – in the younger age groups (0–24 years), followed by a remarkable and quite steady increase thereafter (see Fig. 33). After the age of 64 years, a continued rise in incidence rates for both sexes occurs, the peak being reached in the oldest age group (85 years and over) with mortality rates of 130/100 000 females and 140/100 000 males. However, the data also show that the largest relative differences between males and females actually occur in the middle age range (20–59 years of age), and that this difference almost disappears with rising age.

A national example of fall mortality stratified by age and sex for Romania (see Annex 1) indicates that sex-related differences can be more strongly expressed within certain countries.


Fig. 32. Fall mortality rate/100 000 population by sex (last year of reporting)

0

5

10

15

20

25

Por

tuga

l

Spa

in

Italy

Den

mar

k

Gre

ece

Uni

ted

Kin

gdom

Icel

and

Luxe

mbo

urg

Nor

way

Irela

nd

Fran

ce

Sw

eden

Net

herla

nds

Ger

man

y

Bel

gium

Aus

tria

Sw

itzer

land

Finl

and

Eur

o 1

coun

tries

[a]

All

coun

tries

[a]

(n=4

4)

Mor

talit

y ra

te/1

0000

0

0

1

2

3

4

5

Rat

io o

f mor

talit

y ra

tes

(mal

e:fe

mal

e)

0

5

10

15

20

25

Cyp

rus

Bul

garia

Latv

ia

Rom

ania

Slo

vaki

a

Est

onia

Pol

and

Mal

ta

Cze

ch R

epub

lic

Lith

uani

a

Hun

gary

Slo

veni

a

Eur

o 2

[a]

All

coun

tries

[a]

(n=4

4)

Mor

talit

y ra

te/1

0000

0

0.0

1.5

3.0

4.5

6.0

7.5

Rat

io o

f mor

talit

y ra

tes

(mal

e:fe

mal

e)

0

5

10

15

20

25

Aze

rbai

jan

Geo

rgia

Kyr

gyzs

tan

Uzb

ekis

tan

Kaz

akhs

tan

Rep

ublic

of M

oldo

va

Ukr

aine

Rus

sian

Fed

erat

ion

Bel

arus

Eur

o 3

coun

tries

[a]

Isra

el

Alb

ania

MK

D [b

]

Ser

bia

Cro

atia

Eur

o 4

coun

tries

[a]

All

coun

tries

[a]

(n=4

4)

Mor

talit

y ra

te/1

0000

0

0.0

1.5

3.0

4.5

6.0

7.5

Rat

io o

f mor

talit

y ra

tes

(mal

e:fe

mal

e)

Male Female Sex ratio

Source: data from DMDB, 2011.Notes: [a] average of national rates; [b] MKD: ISO code for the former Yugoslav Republic of Macedonia.

Target groups for actionFatal falls appear to affect males and older people to a greater extent. Globally, sex-related differences are more pronounced in the middle age categories (30–60 years) but remain smaller than age-related differences. Fall injuries among older people are, in fact, a major public health concern in Europe. A variety of individual factors contribute to explain the vulnerability of the older population, including frailty, impaired cognitive and visual capacity to handle certain situations and general health condition (OECD, 2001). In addition, the surrounding environment inside and outside the home can either aggravate or minimize the consequences of those individual vulnerability factors.


Fig. 33. Fall mortality rate/100 000 population by age (last year of reporting)

can the scale be reduced? Values go up to 150 but the scale till 1000…

0

1

10

100

1000<1 1–

4

5–9

10–1

4

15–1

9

20–2

4

25–2

9

30–3

4

35–3

9

40–4

4

45–4

9

50–5

4

55–5

9

60–6

4

65–6

9

70–7

4

75–7

9

80–8

4

85+

Age group

Mor

talit

y ra

te/1

0000

0

Male

Female

Source: data from DMDB, 2011.

Although the data do not provide evidence on the socioeconomic dimension of falls, some evidence from other studies and data sets can be referred to. In 2000 there were strong differences in fatal falls between high-income and low- and middle-income European countries, with 2.4 times more DALYs lost in low- and middle-income countries than high-income countries (Peden, McGee and Krug, 2002).

Some studies have indicated that high SES is associated with a lower rate of falls among children and older people, but the impact of SES does not seem to be consistent (WHO, 2007a; Laflamme, Burrows and Hasselberg, 2009; Sethi et al., 2008). The same applies to studies conducted at the area level, such as those comparing different neighbourhoods.

Health implications77% of all home injuries among small children (up to five years) are due to falls (Bauer and Steiner, 2009), but in childhood, falls are seldom lethal and many children suffer only minor injuries and temporary disabilities due to falls. The most frequent injuries to children from falls are fractures (43%) followed by contusions and bruises (22%) (Kuratorium fur Verkehrssicherheit, 2009). In adolescence and adulthood, falls frequently lead to hospitalization, and the effects include temporary as well as chronic functional constraints or other general limitations (Williams et al., 2011).

Among older people, falls have the most severe health outcomes, such as hip fractures, and show strongly increasing mortality rates from the age of 65 years. Fall-related injuries can impair autonomy and quality of life dramatically; as a result, a fear of falling is frequently observed among older people, restricting physical activity and mobility levels and possibly even increasing the risk of falls. In addition, such fear impacts social life negatively and can even lead to depression (WHO, 2007a).

Conclusions and suggestionsThe available data indicate that age-related differences are more important than sex-related differences when it comes to fall injury mortality. However, sex-related inequalities exist and vary substantially with age: male mortality from falls surpasses that of females at all ages, although the highest difference between the sexes is found among the middle age categories.

In order to reduce the inequality in fall death rates, fatality as the most severe outcome in old age should be prevented and reduced. Efforts should be made to maintain or improve physical capacity through fitness programmes or to adapt dwellings to the needs and abilities of the elderly population through


home modification programmes. Regional inequality data reveal that the reported fall-related mortality rates for older people are surprisingly low in the Euro 3 countries. However, research shows that there were no governmental injury prevention programmes reported or in place in that region (Armour-Marshall et al., 2011), which may indicate a need for more research and monitoring.

Young children would benefit from higher safety standards at home (European Child Safety Alliance, 2009) as they suffer most from fatal falls within the age group below 25 years. Targeted action to protect children could therefore also help to close this gap.

Suggested mitigation actions are:• safer building codes and especially adequate and adaptable homes for the elderly;• a number of evidence-based safe practices to prevent the occurrence and consequences of

children falling in the home (such as window guards and stair gates) (European Child Safety Alliance, 2009);

• creating and maintaining physical activity-friendly residential environments that support active living and especially active ageing;

• better research into the socioeconomic and sociodemographic determinants of fatal falls and the occurrence of falls in adults – especially adult males – to create targeted action to decrease inequalities and avoid an early onset of the risk of fatal falls.

CONCluSION ON INjury-rElATEd INEquAlITIESInjuries are an increasingly important source of concern both globally and within Europe, where they are a leading cause of mortality and disability. Injuries are, however, often preventable, whether sustained at home, at work, in the traffic environment or during leisure time: numerous evidence-based strategies have already been identified that can be implemented to reduce either the likelihood of their occurrence or the severity of their consequences (Sethi et al., 2006b; Peden et al., 2004; WHO, 2007c).

Sociodemographic inequalities in injury rates are also preventable: they can be avoided and they are reversible (Laflamme, 1998). Reaching and sustaining health targets in the WHO European Region requires not only safety-for-all policies and interventions but also equality-oriented ones (Whitehead and Dahlgren, 2006; Dahlgren and Whitehead, 2006). Equality-oriented measures may build on initiatives aimed at narrowing the safety divide between the worse- and better-off or focused solely on people or neighbourhoods in poverty (Laflamme, Burrows and Hasselberg, 2009; WHO, 2009b).

In the foregoing analysis, emphasis was placed on demographic differences within and between European countries and subregions. The differences between individual countries are much more strongly expressed than those between regions, making deductions on a subregional level difficult. Sex-related differences to the detriment of males were reconfirmed, although the magnitude of the differences by sex varied across health inequity indicators. For RTIs, for instance, the highest sex-related inequality arose among young adults (25–29 years), where the male mortality rate was more than five times higher than that of females. For fall-related mortality, rates were on average more than twice higher in males than females, but sex-related differences were most pronounced in the 20–59 year age range. Age-related differences varied too: the most striking ones were found for fall-related mortality, where rates were relatively low during the first decades of life, increasing somewhat thereafter, and rising remarkably after the age of 60, reaching a peak in the eighties. The analysis was based on severe (work-related) and fatal (RTI-, fall-, poisoning-related) injuries, but those represent “the tip of the iceberg”. A set of analyses comparing high-income countries with low- and middle-income ones indicated that a high level of national income is associated with a lower risk of injury. Although this observation


was expected, the mechanisms lying behind it remain to be determined. It is also of note that wide differences exist between countries of similar economic levels, which is why a potential for improvement exists for all countries.

Recommendations include a set of diverse actions and involve resources from different sectors, including – but not limited to – the health sector. The experience accumulated from those sectors where extensive progress has been made, at least in some countries, demonstrates that environmental changes and so-called “passive” measures of prevention have the greatest and most long-lasting impact on injury rates (WHO, 2007c; WHO, 2009); they also have the advantage of being less stigmatizing. What such measures do, using Haddon’s (1980) terminology, is “eliminate”, “separate”, “isolate”, or “modify” the sources of danger. Similar strategies have been employed in various environments (for example, in traffic, at home and at work) with considerable success. Taking the traffic environment as an example, differential exposure to hazards may be addressed by countermeasures ranging from modifications of the environment itself (such as traffic separation or traffic calming) to improvements in the functioning of public transport systems (see Table 7).

Table 7. Haddon matrix applied to a road traffic crash

Phases Factors

Human vehicle Physical environment Social environment

Pre-event Attitudes Knowledge Use of alcohol Driver experience

Vehicle condition Speed

Roadway design Traffic calming Pedestrian facilities

Traffic laws Cultural norms

Event Use of seat belts Wearing fastened helmet

Seat belts Helmets

Shoulders, medians Guardrails

Helmet and seat belt laws

Post-event First aid Medical treatment

Fire risk Availability of trauma care equipment Traffic congestion

Standards of trauma care in hospitals

Source: data from Hazen and Ehiri, 2006, adapted.

It must also be emphasized that prevention and control of injury risk are significantly aided by – and sometimes a pre-condition for – well-defined and enforced legislation and regulations (WHO, 2007c). Such legislation can be safety-oriented, determining minimum standards and conditions under which a number of activities or tasks can be performed (for example, during leisure time and sports or on the road). It can also impose safe behaviours and practices that would not be adopted by all on a voluntary basis only (such as the fitting of car restraints, bicycle helmet use and reducing alcohol consumption). Well-anchored legislation has a great potential not only to improve safety for all but also to narrow the safety divide and reduce injury-related inequality.

rEFErENCES

Armour-Marshall J et al. (2011). Childhood deaths from injuries: trends and inequalities in Europe. European Journal of Public Health (first published online February 9, 2011, DOI: 10.1093/eurpub/ckr004).

Bauer R, Steiner M (2009). Injuries in the European Union: statistics summary 2005–2007. Vienna, Kuratorium fur Verkehrssicherheit (Austrian Road Safety Board).

Beyer FR, Ker K (2009). Street lighting for preventing road traffic injuries. Cochrane Database of Systematic Reviews, Issue 1, Art. No. CD004728 (DOI: 10.1002/14651858.CD004728.pub2).

Bunn F et al. (2003). Area-wide traffic calming for preventing traffic related injuries. Cochrane Database of Systematic Reviews, Issue 1, Art. No. CD003110 (DOI: 10.1002/14651858.CD003110).

Centre for Research and Prevention of Injuries (2011). Fact sheet: prevention of accidental poisoning among elderly. Athens, EUNESE (http://www.euroipn.org/eunese/Documents/FS%20EN/FS_Poisoning.pdf, accessed 21 October 2011).


Costa G, D’errico A (2006). Inequalities in health: do occupational risks matter? European Journal of Public Health, 16(4):340.

Council Directive 89/391/EEC of 12 June 1989 on the introduction of measures to encourage improvements in the safety and health of workers at work (1989). Luxembourg, European Communities (http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:1989:183:0001:0008:EN:PDF, accessed 13 October 2011).

Dahlgren G, Whitehead M (2006). European strategies for tackling social inequities in health: levelling up, part 2. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0018/103824/E89384.pdf, accessed 21 October 2011).

Directorate General for Health and Consumer Protection (2006). Factsheet: alcohol-related harm in Europe – key data. Brussels, European Communities (http://ec.europa.eu/health/archive/ph_determinants/life_style/alcohol/documents/alcohol_factsheet_en.pdf, accessed 13 October 2011).

European Agency for Safety and Health at Work: reducing workplace accidents: advice for employers [web site] (2011) (http://osha.europa.eu/en/topics/accident_prevention/risks, accessed 6 June 2011).

European Child Safety Alliance (2009). Childhood falls. Amsterdam, EuroSafe (http://www.childsafetyeurope.org/publications/info/factsheets/childhood-falls.pdf, accessed 21 October 2011).

European Child Safety Alliance (2011). Poisoning. Amsterdam, EuroSafe (http://www.childsafetyeurope.org/publications/info/factsheets/childhood-poisoning.pdf, accessed 1 June 2011).

European Commission (2007). Improving quality and productivity at work: community strategy 2007–2012 on health and safety at work. Brussels, European Commission (http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2007:0062:FIN:EN:PDF, accessed 9 September 2011)

European Commission (2009). Causes and circumstances of accidents at work in the EU. Luxembourg, European Communities (http://ec.europa.eu/social/main.jsp?catId=738&langId=en&pubId=207&furtherPubs=yes, accessed 21 October 2011).

European Commission (2010). Towards a European area road safety area: policy orientations on road safety 2011–2020. Brussels, European Commission (http://ec.europa.eu/transport/road_safety/pdf/com_20072010_en.pdf, accessed 21 October 2011).

European Commission (2011). Road safety knowledge base. Brussels, European Commission (http://ec.europa.eu/transport/road_safety/specialist/knowledge/index.htm, accessed 28 April 2011).

European detailed mortality database [online database] (2011). July 2010 version. Copenhagen, WHO Regional Office for Europe (http://data.euro.who.int/dmdb/, downloaded on 15 February 2011).

European Health for All database [online database] (2011). January 2011 version. Copenhagen, WHO Regional Office for Europe (http://data.euro.who.int/hfadb, downloaded on 18 April 2011).

European mortality database [online database] (2011). January 2011 version. Copenhagen, WHO Regional Office for Europe (http://data.euro.who.int/hfamdb, downloaded on 19 April 2011).

European Transport and Safety Council (2007). Social and economic consequences of road traffic injury in Europe. Brussels, European Transport and Safety Council (http://www.etsc.eu/documents/Social%20and%20economic%20consequences%20of%20road%20traffic%20injury%20in%20Europe.pdf, accessed 21 October 2011).

Eurostat [online database] (2011). Accidents at work – until 2007 [data set]. Luxembourg, Eurostat (http://epp.eurostat.ec.europa.eu/cache/ITY_SDDS/EN/hsw_acc7_work_esms.htm, downloaded on 06 June 2011).

Haddon W Jr (1980). Advances in the epidemiology of injuries as a basis for public policy. Public Health Reports, 95(5): 411–421 (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1422748/?tool=pmcentrez, accessed 31 October 2011).

Hazen A, Ehiri JE (2006). Road traffic injuries: hidden epidemic in less developed countries. Journal of the National Medical Association, 98(1):73–82.

ILO (1981). Convention concerning occupational safety and health and the working environment (C155). Geneva, International Labour Organization (http://www.ilo.org/ilolex/english/convdisp1.htm, accessed 14 November 2011).

ILO (1985). Convention concerning occupational health services (C161). Geneva, International Labour Organization (http://www.ilo.org/ilolex/english/convdisp1.htm, accessed 14 November 2011).


Kay D et al. (2011). Fairness in a car-dependent society. London, Sustainable Development Commission (http://www.sd-commission.org.uk/data/files/publications/fairness_car_dependant.pdf, accessed 21 October 2011).

Kelsey JL et al. (2010). Indoor and outdoor falls in older adults are different: the maintenance of balance, independent living, intellect, and zest in the elderly of Boston Study. Journal of the American Geriatrics Society, 58(11): 2135–2141.

Kuratorium fur Verkehrssicherheit (ed.) (2009). Data report: injuries in the EU due to falls of children at home. EU Injury Data & Reporting Services. Vienna, Kuratorium fur Verkehrssicherheit (Austrian Road Safety Board) (https://webgate.ec.europa.eu/idb/documents/EU_injury_data_report_2009_Falls-Children-Home.pdf, accessed 31 October 2011).

Laflamme L (1998). Social inequality in injury risks: knowledge accumulated and strategies for the future. Stockholm, Karolinska Institute and National Institute of Public Health (Report 1998: 33).

Laflamme L, Diderichsen F (2000). Social differences in traffic injury risks in childhood and youth-a literature review and research agenda. Injury Prevention, 6:293–298.

Laflamme L, Menckel E (1995). Aging and occupational accidents. A review of the literature of the last three decades. Safety Science, 21(2):145–161.

Laflamme L, Burrows S, Hasselberg M (2009). Socioeconomic differences in injury risks: a review of findings and a discussion of potential countermeasures. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0012/111036/E91823.pdf, accessed 21 October 2011).

Laflamme L, Hasselberg M, Burrows S (2010). 20 years of research on socioeconomic inequality and children’s-unintentional injuries understanding the cause-specific evidence at hand. International Journal of Pediatrics, Issue 2010, Art. No. 819687, DOI: 10.1155/2010/819687 (http://www.hindawi.com/journals/ijped/2010/819687, accessed 21 October 2011).

Laflamme L et al. (2009). Addressing the socioeconomic safety divide: a policy briefing. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0004/96457/E92197.pdf, accessed 21 October 2011).

Mock C et al. (2004). Strengthening the prevention and care of injuries worldwide. Lancet, 363(9427):2172–2179.

OECD (2001). Ageing and transport: mobility needs and safety issues. Paris, OECD (http://www.oecd.org/dataoecd/40/63/2675189.pdf, accessed 21 October 2011).

Peden M, McGee K, Krug E, eds. (2002). Injury: a leading cause of the global burden of disease, 2000. Geneva, World Health Organization (http://whqlibdoc.who.int/publications/2002/9241562323.pdf, accessed 13 October 2011).

Peden M et al. (2004). World report on road traffic injury prevention. Geneva, World Health Organization (http://whqlibdoc.who.int/publications/2004/9241562609.pdf, accessed 28 April 2011).

Pruss-Üstun A, Corvalán C (2006). Preventing disease through healthy environments. Towards an estimate of the environmental burden of disease. Geneva, World Health Organization (http://www.who.int/quantifying_ehimpacts/publications/preventingdisease.pdf, accessed 13 October 2011).

Racioppi F et al. (2004). Preventing road traffic injury: a public health perspective for Europe. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0003/87564/E82659.pdf, accessed 13 October 2011).

Sethi D, Mitis F, Racioppi F (2010). Preventing injuries in Europe: from international collaboration to local implementation. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0011/96455/E93567.pdf, accessed 21 October 2011).

Sethi D et al. (2006a). Injuries and violence in Europe: why they matter and what can be done. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0005/98762/E88037.pdf, accessed 21 October 2011).

Sethi D et al. (2006b). Reducing inequalities from injuries in Europe. Lancet, 386(9554):2243–2250.

Sethi D et al. (2008). European report on child injury prevention. Copenhagen, WHO Regional Office for Europe (http://www.who.int/violence_injury_prevention/child/injury/world_report/European_report.pdf, accessed 13 October 2011).


Siegrist J et al. (2010). Employment arrangements, work conditions and health inequalities. Task group report to the Marmot Review. London, University College London.

Stirbu I et al. (2006). Injury mortality among ethnic minority groups in the Netherlands. Journal of Epidemiology and Community Health, 60(3):249–255.

Thomson H (2011). Housing improvements and their health effects. In: Braubach M, Jacobs D, Ormandy D, eds. Environmental burden of disease associated with inadequate housing: methods for quantifying health impacts of selected housing risks in the WHO European Region. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0003/142077/e95004.pdf, accessed 7 November 2011).

Tuchsen F et al. (2009). Work injuries and disability. Journal of Safety Research, 40(1):21–24.

Whitehead M, Dahlgren G (2006). Concepts and principles for tackling social inequities in health: levelling up, part 1. Copenhagen, WHO Regional Office for Europe (http://www.enothe.eu/cop/docs/concepts_and_principles.pdf, accessed 21 October 2011).

WHO (2002). The World health report 2002 – reducing risks, promoting healthy life. Geneva, World Health Organization (http://www.who.int/whr/2002/en, accessed 25 May 2008).

WHO (2007a). WHO global report on falls prevention in older age. Geneva, World Health Organization (http://www.who.int/ageing/publications/Falls_prevention7March.pdf, accessed 13 October 2011).

WHO (2007b). Workers’ health: global plan of action. Sixtieth World Health Assembly, 2007. Geneva, World Health Organization (http://apps.who.int/gb/ebwha/pdf_files/WHA60/A60_R26-en.pdf, accessed 19 September 2011).

WHO (2007c). Preventing injuries and violence: a guide for Ministries of Health. Geneva, World Health Organization (http://whqlibdoc.who.int/publications/2007/9789241595254_eng.pdf, accessed 21 October 2011).

WHO (2009). European status report on road safety: towards safer roads and healthier transport choices. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0015/43314/E92789.pdf, accessed 21 October 2011).

WHO (2010a). International statistical classification of diseases and related health problems 10th revision (ICD-10): transport accidents. Geneva, World Health Organization (http://apps.who.int/classifications/icd10/browse/2010/en#/V01-V99, accessed 21 October 2011).

WHO (2010b). Social and gender inequalities in environment and health. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0010/76519/Parma_EH_Conf_pb1.pdf, accessed 13 October 2011).

WHO (2010c). Global plan for the decade of action for road safety 2011–2020. Geneva, World Health Organization (http://www.who.int/roadsafety/decade_of_action/plan/english.pdf, accessed 13 October 2011).

WHO (2010d). Global Strategy to reduce the harmful use of alcohol. Geneva, World Health Organization (http://www.who.int/substance_abuse/alcstratenglishfinal.pdf, accessed 4 September 2011).

WHO (2010e). Falls: fact sheet no. 344. Geneva, World Health Organization (http://www.who.int/mediacentre/factsheets/fs344/en/, accessed 10 May 2011).

WHO (2011). Global burden of disease: disease and injury regional mortality estimates for 2008. Geneva, World Health Organization (http://www.who.int/healthinfo/global_burden_disease/estimates_regional/en/index.html, accessed 21 June 2011).

Williams J et al. (2011). Longer term health of young and middle-aged adults following unintentional falls at home resulting in hospitalization. Injury (first published online April 14, 2011, DOI: 10.1016/j.injury.2011.03.050).

Winter Griffith H, Moore S, Yoder K (2006). Complete guide to symptoms, illness and surgery, 5th ed. London, Penguin books.

4

ENVIRONMENT-RELATED INEQUALITIES

CHAPTER

Chapter 4. Environment-related inequalities86

CHAPTEr 4. ENvIrONmENT-rElATEd INEquAlITIESGabriele Bolte

Key messages

• In EU15 countries, prevalence of complaints about noise from neighbours or from the street is higher among individuals living in relative poverty. This inequality is not apparent in NMS12 countries.

• The direction and degree of social inequalities in complaints about lack of access to recreational or green areas depends on the local or regional situation in a given country or region, and on the socioeconomic indicator analysed. The expectation that disadvantaged groups might be more affected by a lack of access to recreational or green areas is mostly met in the EU15 region, while NMS12 and Euro 4 countries show more variation.

• Gender differences are relevant in sociodemographic inequalities in access to recreational or green areas.

• Among non-smokers, socioeconomic inequalities in second-hand smoke exposure at home exist with a higher exposure among disadvantaged groups, characterized by unemployment, difficulty paying bills most of the time, or low self-assessed social position.

• Among non-smokers in EU27 countries, prevalence of exposure to second-hand smoke indoors at work is higher among manual workers and among those individuals with difficulty paying bills most of the time.

• In the majority of countries, working non-smoking males were more often exposed to second-hand smoke at work than working non-smoking females.

• In accordance with the scientific literature, independent of the socioeconomic indicator used, socially disadvantaged individuals in EU15 countries mostly have a higher self-assessed exposure to noise, lack of access to recreational or green areas, and exposure to second-hand smoke at home or at work. In contrast, socially advantaged individuals in NMS12 countries appear to reside in urban neighbourhoods where environmental burdens such as noise or lack of access to recreational or green areas may occur more frequently.


INTrOduCTIONIt has been estimated that approximately one-quarter of the global disease burden – and more than one-third of the burden among children – is caused by modifiable environmental factors, which include physical, chemical and biological hazards. Cardiovascular disease and cancer, as well as health-related behaviour – such as physical activity as a mediating factor – are the most relevant outcomes of noncommunicable disease within this estimation of the environmental disease burden in developed countries (Pruss-Üstun and Corvalán, 2006). A pilot project on the environmental burden of disease in Europe showed that 3–7% of the standard WHO discounted age-weighted burden of disease was associated with nine selected environmental stressors in the six participating countries. Among these nine stressors air pollution had the highest public health impact, followed by second-hand smoke and traffic noise (Hänninen and Knol, 2011).

There is growing evidence that socioeconomic inequalities in man-made environmental conditions contribute to health inequalities (Evans and Kantrowitz, 2002; Brulle and Pellow, 2006; Gee and Payne-Sturges, 2004). Exposure to environmental hazards as well as access to environmental benefits may differ according to socioeconomic position. The final report of the CSDH also emphasizes the relevance of poor and unequal living conditions for health inequalities, considering environmental factors in the physical form of the built environment, the quality of the natural environment in which people reside, and the nature of employment and working conditions for health (CSDH, 2008).

In addition, socioeconomic factors may modify health impacts by influencing an individual’s vulnerability (WHO, 2010; Bolte, Pauli and Hornberg, 2011). Psychosocial stress has been suggested as a vulnerability factor linking social conditions with environmental hazards (Gee and Payne-Sturges, 2004).

Most research on socioeconomic disparities in environmental exposures has focused on environmental hazards and pollutants, showing, for example, that some disadvantaged urban subpopulations have a higher exposure to ambient air pollution, with motor vehicle traffic as one important source; that areas of high exposure often coincide with a lower socioeconomic position; and that socially disadvantaged people may be more responsive to air pollution (Kinney and O’Neill, 2006; Laurent et al., 2007). Studies within Europe indicate that less affluent population groups are most exposed to environmental risks in their place of residence and that waste facilities are often disproportionately located in areas with more deprived residents (Braubach and Fairburn, 2010; Martuzzi, Mitis and Forastiere, 2010). Recently, there have been more attempts to consider environmental resources – such as access to public green areas – and the distribution of these amenities in urban neighbourhoods, which may influence health-related behaviour, physical and mental health. Socioeconomic differences in exposure to ambient air pollution, noise, second-hand smoke and lack of access to green spaces have repeatedly been shown to exist especially among children in Europe (Bolte and Kohlhuber, 2005; Bolte, Tamburlini and Kohlhuber, 2010).

The aim of this chapter on environment-related inequalities is to illustrate socioeconomic differences in environmental exposures related to noise exposure at home, lack of access to recreational or green areas, and exposure to tobacco smoke at home and at work. Using available data from cross-national surveys of the WHO European Region, this chapter focuses on social indicators of material circumstances (such as income and poverty) according to the international debate on environmental inequalities and environmental justice (Brulle and Pellow, 2006; O’Neill et al., 2007).


dATA ANd mETHOdS

Noise exposure at homeData were retrieved from the EU-SILC database for the years 2004 to 2009. Data were available for between 15 (2004) and 30 countries (2009: 27 EU Member States, Iceland, Norway and Switzerland). Details of the sample size of the cross-sectional studies for each country are given on the EU-SILC web site.8 The reference population for EU-SILC includes all private households. Household members aged 16 or over were interviewed.

Noise from neighbours or from the street was assessed based on the EU-SILC question “Do you have any of the following problems related to the place where you live?” including answer category “Too much noise in your dwelling from neighbours or from outside (traffic, business, factory, etc.)” with answer options Yes and No. The self-reported complaints about noise from neighbours or from the street are given as percentage of the total study population of each country.

No stratification by sex was possible. Relative poverty was defined as a single-adult equivalent disposable income attributed to all household members below 60% of national median equivalized income, the variable obtained from the EU-SILC database. For equivalization of the total disposable household income the OECD modified scale of equivalization factors was used. Categories of household type “Households with dependent children” and “Households without dependent children” were used. The category “Single parent with dependent children” was also considered.

As the data provided cover the EU countries and Iceland, Norway and Switzerland only, EU15 and NMS12 were chosen as subregions, as with the other Eurostat-based inequality indicators in the injury- and housing-related inequality chapters. Data for Iceland, Norway and Switzerland are thus presented separately.

lack of access to recreational or green areasThe second European Quality of Life Survey (EQLS) was conducted by Eurofound in 2007 with more than 35 000 adult interviewees from 31 countries: the 27 EU Member States, Norway, and three EU candidate countries (Croatia, the former Yugoslav Republic of Macedonia and Turkey). Data were obtained from the web site of the Economic and Social Data Service, via the UK Data Archive of the University of Essex. The dataset comprised data for 35 634 people (57% female).

EQLS participants were asked “Please think about the area where you live now – I mean the immediate neighbourhood of your home. Do you have very many reasons, many reasons, a few reasons, or no reason at all to complain about each of the following problems?” including answer category “Lack of access to recreational or green areas”. For analysis, the answers to this question on reasons to complain were reduced to three categories: Yes (“Very many reasons”, “Many reasons” and “A few reasons”), No (“No reason at all”) and Don’t know.

Data for several socioeconomic indicators were used for the inequality assessment.• Income: the dataset comprised a variable of OECD income, with four categories – Lowest, Low,

High, and Highest – indicating quartiles of income.• Difficulty paying bills: participants were asked, “A household may have different sources

of income and more than one household member may contribute to it. Thinking of your household’s total monthly income: is your household able to make ends meet…?” The answer options were “Very easily”, “Easily”, “Fairly easily”, “With some difficulty”, “With difficulty”, “With great difficulty”, and “Don’t know”. For analysis, the answers were reduced to three categories:

8 See the Eurostat web site (http://epp�eurostat�ec�europa�eu/portal/page/portal/income_social_inclusion_living_conditions/documents/tab/Tab/EU-SILC%20sample%20size�pdf)�


º No difficulty paying bills (“Very easily”, “Easily”, “Fairly easily”) º Difficulty paying bills from time to time (“With some difficulty”) º Difficulty paying bills most of the time (“With difficulty”, “With great difficulty”).

• Employment: the available information was summarized into three categories – Employed (including self-employed), Unemployed and Other (including home-maker and retired).

• Level of education: participants were asked “What is the highest level of education you completed?” The answers were reduced to three categories:

º Low (“no education completed”) º Medium (“primary education”, “lower secondary education”, or “upper secondary education”) º High (“post-secondary, including pre-vocational or vocational education but not tertiary”, “tertiary education – first level”, or “tertiary education – advanced level”).

• Household type: this variable differentiates four categories – Single person, Single parent, Couple without children and Couple with children.

The socioeconomic indicators given above were only moderately correlated. All analyses were stratified by sex.

As the data provided cover the EU countries and Norway as well as the three candidate countries, subregions used to present results for geopolitical areas were EU15, NMS12, and Euro 4 countries; data for Norway are thus presented separately in the figures.

Second-hand smoke exposure at home or at workData on the second-hand smoke exposure at home and at work indicators were derived from Special Eurobarometer 332: tobacco (TNS Opinion and Social, 2010) and obtained through the WHO European Centre for Environment and Health. Special Eurobarometer 332: tobacco was part of wave 72.3 of the Eurobarometer and was conducted in 2009. It comprised 30 292 residents (56% female) aged 15 or over. The analysis was restricted to 29 792 participants (56% female) of 30 countries (27 EU Member States and three candidate countries: Croatia, the former Yugoslav Republic of Macedonia and Turkey).

To assess potential second-hand smoke exposure at home, the participants were asked “Which statement best describes smoking situation inside your house?” with answer options:• “Smoking is not allowed at all inside the house”• “Smoking is allowed only in certain rooms inside the house”• “Smoking is allowed everywhere inside the house”.

For analysis, potential second-hand smoke exposure at home was defined as “Smoking is allowed only in certain rooms or everywhere inside the house”.

To assess exposure to second-hand smoke at work, only those participants currently working were asked “How often are you exposed to tobacco smoke indoors at your workplace?” with answer options:• “Never or almost never”• “Less than 1 hour a day”• “1–5 hours a day”• “More than 5 hours a day”• “Not relevant (don’t work or don’t work indoors)”.

Since the majority of participants answered “Never or almost never”, exposure to second-hand smoke at work was defined as any exposure, combining categories “Less than 1 hour a day”, “1–5 hours a day”, and “More than 5 hours a day” for analysis. In most countries the proportion of individuals answering that they do not work indoors was below 10% of all non-smokers; exceptions were Turkey (19%) and Greece (12%).


To assess whether the lowest exposure category “Less than 1 hour a day” might cause artificial results, a sensitivity analysis was carried out defining exposure to second-hand smoke at work as at least 1 hour a day.

Several indicators of socioeconomic position were used for the inequality assessment.• Self-assessed position on the social scale: participants were asked “On the following scale, step ‘1’

corresponds to the lowest level in the society; step ‘10’ corresponds to the highest level in the society. Could you tell me on which step you would place yourself?” This position was classified into Low (steps 1–4), Medium (steps 5–6), and High (steps 7–10) in the dataset.

• Difficulty paying bills: the question was “During the last twelve months, would you say you had difficulties to pay your bills at the end of the month?” Answer options were “Most of the time”, “From time to time” and “Almost never or never”. A fourth category comprised those participants who refused to answer this question.

• Employment (applied for exposure at home). The available information was summarized into three categories:

º Employed (including self-employed, manager, other white collar worker and manual worker); º Unemployed; º Other (including home-maker, retired and student).

• Occupation (applied for exposure at work): data were stratified into two categories of Manual workers and Managers.

No information on household income or level of education was available in this dataset. All analyses were stratified by sex.

All analyses of second-hand smoke exposure were restricted to non-smokers. To assess their current smoking status, participants were asked: “Regarding smoking cigarettes, cigars or a pipe, which of the following applies to you?” with answer options “Smoke at the present time”, “Used to smoke but have stopped” and “Never smoked”. The smoking status of respondents was classified as current smoker or non-smoker (“Never smoked” or “Used to smoke but have stopped”). Depending on the country, 58% to 84% of the respondents were non-smokers (males: 49–88%, females: 62–85%). In total in EU27, 72% were non-smokers (males: 66%, females: 77%).

As the Eurobarometer data covered the EU countries as well as the three candidate countries, subregional groupings used were EU15, NMS12, and Euro 4 countries.

When considering social inequalities in second-hand smoke exposure, it must be borne in mind that current smokers mostly have a lower socioeconomic position (see Table 8), so restricting the analysis to non-smokers causes a disproportionate loss of socially disadvantaged individuals and may result in an underestimation of inequalities. Nevertheless, from a health point of view, second-hand smoke exposure is more relevant in non-smokers than in active smokers.

Among non-smokers, socioeconomic indicators of self-assessed social position, difficulty paying bills, and employment were only weakly correlated.

For the indicators on lack of access to green space and exposure to second-hand smoke, it proved to be very difficult – and sometimes impossible – to generate representative subregional results due to a lack of adequate data. Therefore, the subregional data often represent the average of the national rates of the countries covered by the respective region. In each figure, this restriction is clearly marked as the average of national rates for all reporting countries of the subregion. Subregional data that are representative (often provided by Eurostat databases, as in the case of noise exposure) do not include this indication and the aggregated results are representative of the respective subregion.


Table 8. Socioeconomic characteristics of smokers and non-smokers by sex (%)

male Female

Smoker Non-smoker Smoker Non-smoker

Eu27Employment: unemployed 11�1 4�5 10�3 4�6Self-assessed social position: low 24�3 17�4 22�2 20�3Difficulty paying bills: most of the time 12�4 6�0 15�5 8�8Eu15Employment: unemployed 9�5 4�3 9�3 4�1Self-assessed social position: low 19�0 12�1 17�2 13�5Difficulty paying bills: most of the time 9�6 4�1 11�9 6�0NmS12Employment: unemployed 13�4 4�8 11�5 5�3Self-assessed social position: low 31�1 24�1 28�5 28�9Difficulty paying bills: most of the time 15�9 8�4 20�0 12�2Euro 4 countriesEmployment: unemployed 20�9 12�2 19�2 12�2Self-assessed social position: low 35�0 32�6 32�9 32�2Difficulty paying bills: most of the time 24�7 17�7 28�1 20�4

Source: data from Special Eurobarometer 332: tobacco (TNS Opinion and Social, 2010).

dATA rESTrICTIONS ANd lImITATIONSAssessment of noise exposure using EU-SILC data was based on self-reported noise annoyance in terms of complaints about noise from neighbours or from the street. However, noise perception may differ by social position; for example, if less exposed individuals of a higher social position overrate their noise exposure, then the true relationship between social position and noise exposure could be underestimated. A further limitation is that there was no quantification of the self-reported noise exposure. EU-SILC data comprise only two income categories, and because of restricted data access no data on sex were available. Finally, different sources of noise (noise from neighbours and traffic noise) were combined into one question in EU-SILC. From a health perspective these sources should be assessed separately, because different exposure–response relationships may exist for different kinds of noise.

Complaints about lack of access to recreational or green areas is a rather vague exposure indicator. To assess potential health impacts of this characteristic of the built environment, information on the quality, dimension, ease and convenience of access, perception, and usage of the green space or recreational area is necessary (Mitchell, Astell-Burt and Richardson, 2011; Lee and Maheswaran, 2011; Richardson and Mitchell, 2010). Self-reported data should be combined with objective data on at least the location and dimension of green space or recreational area, and green spaces and recreational areas should be distinguished, because there might be different mechanisms linking these environmental characteristics to health. Furthermore, both the presence and the quality of such areas should be distinguished.

The analysis of second-hand smoke exposure at home was based on the survey answer “Smoking is allowed in certain rooms or everywhere inside the house”. Permission to smoke only provides an indication of exposure; it does not enable a full determination of whether residents have actually been exposed to indoor smoke. Moreover, there was no supplementary information, such as the amount smoked at home by any person or the duration of smoking, in order to quantify exposure at home. The validity of self-reported data on active and passive smoking has been questioned. Public debate about smoke-free legislation in some countries may have contributed to a social preference bias towards reporting smoke-free homes, irrespective of actual smoking habits.

The analysis of second-hand smoke exposure at work was also based on self-reported data. Data were provided on duration of exposure, but do not allow any further quantification of exposure. There was no information on the validity of self-reported data on passive smoking at work.


Both analyses were restricted to non-smokers, since exposure to second-hand smoke was assessed to be more health-relevant in non-smokers. Thus, since smoking is more prevalent among socially disadvantaged individuals, this restriction will have caused a disproportionate loss of data on socially disadvantaged individuals.

INEquAlITIES IN NOISE ExPOSurE AT HOmE

IntroductionEnvironmental noise (defined as noise emitted from all sources except industrial workplaces) is an important public health problem. The main exposure is road traffic noise. Sleep disturbance and annoyance, mostly related to road traffic noise, are the key health issues. At least one million healthy life years are lost every year from traffic-related noise in western European countries, including the EU Member States (WHO, 2011). The effects of neighbourhood noise and leisure noise are not addressed in this risk assessment because of a lack of data. Moreover, differences in exposure by sex and among varying socioeconomic groups, as well as additional burdens among potentially vulnerable subgroups are not considered.

Epidemiological studies show, in accordance with the fact that socially disadvantaged people are more likely to live near busy roads, that noise annoyance due to traffic is often higher in people with a lower socioeconomic position. In addition, social inequalities in objectively assessed noise exposure have been demonstrated (Bolte, Tamburlini and Kohlhuber, 2010). However, as the fact sheet on traffic noise exposure from the Netherlands shows, it must be borne in mind that different transport types (such as road, train and air) may have different inequality profiles (see Annex 1).

Indicator analysis: inequalities by income and household typeThe overall prevalence of complaints about noise from neighbours or from the street varies by country between 10% and 35%, with an average of 22% across EU27 (2009 figures).

In the majority of the 30 reporting countries, self-reported noise exposure at home is higher among individuals living in relative poverty, although in six countries this pattern is reversed (see Fig. 34). Prevalence in EU27 is 25% among individuals below the relative poverty threshold and 22% among those above it. However, when the countries are grouped into subregions, the prevalence difference between the two income categories is present in EU15 countries (26% versus 22%) but disappears in NMS12 countries (21% for both income categories).

There is a small but constant difference of 3–4% between the income groups in EU15 countries from 2005 to 2009, showing no reduction in the existing inequality (see Fig. 35). In contrast, there are no clear and consistent differences between income groups in NMS12 countries. Prevalence of noise exposure at home is somewhat higher in EU15 countries, regardless of income. The highest prevalence is observed among individuals living in relative poverty in EU15 countries.

To assess the combined effect of several social dimensions, data on single parents living in relative poverty were contrasted with data on households without dependent children and with a household income above the relative poverty level. In the group of single parents in relative poverty, noise exposure at home varies between 9% and 41% by country, and in the latter group it varies between 11% and 38%. In the majority of countries (21 of 30) the prevalence of noise exposure at home is higher among single-parent households in relative poverty. Across EU27, the prevalence of noise exposure at home is 32% among single-parent households in relative poverty and 23% among households with no children and an income above the relative poverty level, and if the subregional groupings of EU15 and NMS12 are applied, the prevalence difference between the two types of household are more pronounced in the EU15 countries (33% versus 23%) than the NMS12 countries (24% versus 22%).


Fig. 34. Prevalence of complaints about noise from neighbours or from the street by relative poverty level (2009)

0

5

10

15

20

25

30

35

40Ire

land

Swed

en

Gre

ece

Finl

and

Por

tuga

l

Uni

ted

King

dom

Fran

ce

Spai

n

Aus

tria

Belg

ium

Italy

Den

mar

k

Luxe

mbo

urg

Ger

man

y

Net

herla

nds

EU

15

Est

onia

Lith

uani

a

Bulg

aria

Hun

gary

Pola

nd

Slov

enia

Latv

ia

Slo

vaki

a

Cze

ch R

epub

lic

Rom

ania

Mal

ta

Cyp

rus

NM

S 1

2

EU

27

Icel

and

Nor

way

Sw

itzer

land

Pre

vale

nce

(%)

Below relative poverty levelAbove relative poverty level


Fig. 35. Time trend of prevalence of complaints about noise from neighbours or from the street by household income

15

18

21

24

27

30

2005 2006 2007 2008 2009

Pre

vale

nce

(%)

EU15: below relative poverty levelEU15: above relative poverty levelNMS12: below relative poverty levelNMS12: above relative poverty level

Source: data from EU-SILC, 2011.Note: NMS12 figures for 2005/2006 exclude Romania due to lack of data.

Target groups for actionBased on the limited available information, the most exposed groups in terms of the highest prevalence of complaints about noise from neighbours or from the street are low-income groups, especially single parents, mainly in EU15 countries.


Health implicationsThe environmental burden of disease due to environmental noise has been recently estimated for western European countries with a range of 1.0–1.6 million DALYs lost across all health outcomes (WHO, 2011). The estimates are 61 000 DALYs for ischaemic heart disease, 45 000 for cognitive impairment of children, 903 000 for sleep disturbance, 22 000 for tinnitus, and 587 000 for annoyance.

Data from EU-SILC show a consistent but rather small prevalence difference relating to income in noise exposure – defined as complaints about noise from neighbours or from the street – especially in Euro 1 countries. However, the given categorization of the two income groups (below and above 60% of median income) may mask the real extent of social differences across the entire income spectrum. Differences in exposure prevalence can rise to more than 10% when more specific social groups are identified by combining different socioeconomic characteristics (such as single parents with low income).

Since complaints about noise from neighbours or from the street may also be interpreted as a health effect in terms of annoyance, the EU-SILC data already show clear social inequalities in health related to noise exposure. However, from a health point of view it is important to recognize that there is considerable variation in exposure prevalence – or rather in prevalence of annoyance due to noise from external sources (neighbours or street) – between countries, affecting up to 35% of the population.

Conclusions and suggestionsPrevalence of self-reported noise annoyance varies considerably between countries. Irrespective of social differences, in many countries a relevant proportion of the population is affected by noise from neighbours or from the street. In Euro 1 but not Euro 2 countries, prevalence of complaints about noise from neighbours or from the street is higher among individuals with low income. If data were able to be stratified by more categories of income, the social differences seen would probably be greater.

It has been shown that health impacts such as annoyance and sleep disturbance are mostly related to traffic noise. Therefore, data on subjective noise exposure should be gathered separately for different sources of noise. Since more consistent data are available on social inequalities in exposure to traffic-related air pollution, a clearer picture of noise inequalities could be created if data were available for exposure to traffic-related noise only. Wherever possible, measured data on noise exposure should be used to support the results presented on social inequalities in self-reported noise exposure.

Suggested mitigation actions are:• further enforcement of the EU Environmental Noise Directive (Directive, 2002) to tackle the

important public health issue of traffic-related noise;• ensuring that action plans to address noise issues on a regional level take potential social

inequalities in noise exposure into account;• better reporting of objective traffic-related noise exposure and subjective noise annoyance by

sex and socioeconomic group;• better research on combined exposure to noise and air pollutants, including their distribution

by sex and sociodemographic group, for a comprehensive health risk assessment.


INEquAlITIES IN lACK OF ACCESS TO rECrEATIONAl Or GrEEN ArEAS

IntroductionGreen space may influence physical and mental health as well as well-being (Maas et al., 2006; Mitchell and Popham, 2008; Abraham, Sommerhalder and Abel, 2010). It has been proposed that green space – in terms of contact with nature – has an influence on health through restorative properties (Mitchell, Astell-Burt and Richardson, 2011). Both design of the built environment and accessibility of the natural environment have an impact on physical activity and active living (Edwards and Tsouros, 2006).

Data from Europe indicate that socially disadvantaged people often live in places with less access to public green space (Kruize et al., 2007; Bolte, Tamburlini and Kohlhuber, 2010). Socioeconomically deprived areas in Britain, for example, have fewer large green spaces, which are assumed to be more important for health effects than small ones (Mitchell, Astell-Burt and Richardson, 2011). Even where there is equal access to green space, irrespective of social position, the quality of green space may differ, depending on the social characteristics of the neighbourhood (Bolte, Pauli and Hornberg, 2011). In addition to this exposure variation by social group, a study from the Netherlands shows that the percentage of green space in people’s living environment is positively associated with the perceived general health of residents, and that less-educated groups are more sensitive to the characteristics of their physical environmental (Maas et al., 2006). Another British study demonstrates that health inequalities related to income deprivation are lower in populations who live in the greenest areas compared to those who have less exposure to green space (Mitchell and Popham, 2008). Finally, gender differences in the relationship between green space and health have been reported, which may be due to gender differences in the perception and usage of green spaces (Richardson and Mitchell, 2010).

Indicator analysis: inequalities by sex, income, difficulty paying bills, employment, education level and household typePrevalence of complaints about lack of access to recreational or green areas varies between 6% and 67% of EQLS participants by country, and the intercountry prevalence range is comparable between females (7–67%) and males (5–67%). Across EU27, 34% of participants (35% of females, 34% of males) have complaints about this lack of access. When divided into subregions, the prevalence of complaints is 32% (for both males and females) in EU15 countries, 44% (44% of females, 43% of males) in NMS12 countries, and reaches 47% (47% of females, 46% of males) in the three Euro 4 countries – Croatia, the former Yugoslav Republic of Macedonia and Turkey. In Norway the prevalence is 9% among both females and males. In the majority of countries, the highest rate of complaints was found in the 18–34 year age group and the lowest among those 65 years and over.

A pronounced variation in the prevalence difference across the four income groups is observable in many countries (see Fig. 36). Nevertheless, looking at the total for all EU27 countries, there is only a minor difference in the lack of access to recreational or green areas (a maximum difference of 4% between income quartiles for females and males). This is because those individuals in the lowest income quartile in EU15 countries complain of a lack of access more frequently (lowest income quartile: 32% of females and males; highest income quartile: 27% of females, 28% of males) while, in contrast, individuals in the lowest and low income quartile in NMS12 countries complain less often than those in the high and highest (lowest income quartile: 34% of females, 36% of males; highest income quartile: 46% of females, 41% of males). Looking at the three Euro 4 countries together, there are no major social differences between the two extreme income quartiles (lowest income: 51% of females, 46% of males; highest income: 51% of females, 47% of males), although the prevalence of complaints about lack of access to recreational or green areas is lower in the two middle income quartiles for both females and males. However, in Croatia the highest rate of lack of access is observed among individuals in the highest income quartile, in contrast to the former Yugoslav Republic of Macedonia and Turkey where the highest rate is found among individuals in the lowest income quartile.


Fig. 36. Prevalence of complaints about lack of access to recreational or green areas by income quartile and sex (2007)

0

10

20

30

40

50

60

70

80

Den

mar

kFi

nlan

dSw

eden

Net

herla

nds

Uni

ted

King

dom

Fran

ceG

erm

any

Luxe

mbo

urg

Irela

ndA

ustri

aG

reec

eBe

lgiu

mP

ortu

gal

Spai

nIta

lyE

U 1

5 [a

]

Slov

enia

Est

onia

Rom

ania

Cze

ch R

epub

licLi

thua

nia

Hun

gary

Latv

iaS

lova

kia

Cyp

rus

Pola

ndBu

lgar

iaM

alta

NM

S 1

2 [a

]

EU

27

[a]

Cro

atia

Turk

eyM

KD

[b]

Eur

o 4

coun

tries

[a]

Nor

way

Pre

vale

nce

(%)

Lowest quartile Low quartile High quartile Highest quartile

Males

0

10

20

30

40

50

60

70

80

Finl

and

Den

mar

kSw

eden

Aus

tria

Net

herla

nds

Uni

ted

King

dom

Ger

man

yLu

xem

bour

gFr

ance

Irela

ndBe

lgiu

mG

reec

eP

ortu

gal

Spai

nIta

lyE

U 1

5 [a

]

Slov

enia

Est

onia

Rom

ania

Cze

ch R

epub

licH

unga

ryLa

tvia

Slo

vaki

aC

ypru

sPo

land

Lith

uani

aM

alta

Bulg

aria

NM

S 1

2 [a

]

EU

27

[a]

Cro

atia

Turk

eyM

KD

[b]

Eur

o 4

coun

tries

[a]

Nor

way

Pre

vale

nce

(%)

Females

Source: data from EQLS, 2007.Notes: [a] average of national rates; [b] MKD: ISO code for the former Yugoslav Republic of Macedonia.

Stratification of the data by difficulty paying bills most of the time, from time to time or never reveals marked differences in lack of access to recreational or green areas. There is a clear social gradient, with the highest level of disadvantage most often found among those with difficulty paying bills most of the time (see Table 9). Exceptions are found for females in NMS12 countries and males in Euro 4 countries.

National data (see Fig. 37) show that there are large differences between countries regarding both the overall dimension of lack of access to recreational or green areas and the inequalities between males and females with various levels of difficulty paying bills. Several countries show a strong disadvantage for individuals reporting difficulty paying bills, while other countries show a balance or the opposite trend. In several countries, such as the Netherlands, Spain, Estonia, Latvia and the former Yugoslav Republic of Macedonia, variations by sex on the impact of difficulty paying bills can be observed.


Table 9. Lack of access to recreational or green areas by difficulty paying bills (2007)

Female male

No difficulty

difficulty from time to time

difficulty most of the time

No difficulty



EU15 [a] 27�7 33�4 37�2 26�7 34�3 35�2NMS12 [a] 40�5 41�2 38�7 37�7 36�8 40�8EU27 [a] 32�4 35�9 36�8 30�7 34�5 36�9Euro 4 countries [a]

44�8 44�3 50�8 49�3 38�8 46�4

Source: data from EQLS, 2007. Note: [a] average of national rates.

Fig. 37. Prevalence of complaints about lack of access to recreational or green areas by difficulty paying bills and sex (2007)

0102030405060708090

Finl

and

Den

mar

kSw

eden

Net

herla

nds

Ger

man

yU

nite

d Ki

ngdo

mFr

ance

Aus

tria

Irela

ndBe

lgiu

mLu

xem

bour

gSp

ain

Gre

ece

Por

tuga

lIta

lyE

U 1

5 [a

]

Slov

enia

Est

onia

Rom

ania

Cze

ch R

epub

licLa

tvia

Cyp

rus

Slo

vaki

aH

unga

ryPo

land

Lith

uani

aBu

lgar

iaM

alta

NM

S 1

2 [a

]

EU

27

[a]

Cro

atia

Turk

eyM

KD

[b]

Eur

o 4

coun

tries

[a]

Nor

way

Pre

vale

nce

(%)

Females

0102030405060708090

Finl

and

Den

mar

kN

ethe

rland

sSw

eden

Uni

ted

King

dom

Spai

nLu

xem

bour

gFr

ance

Aus

tria

Ger

man

yBe

lgiu

mP

ortu

gal

Gre

ece

Irela

ndIta

lyE

U 1

5 [a

]

Slov

enia

Est

onia

Cze

ch R

epub

licR

oman

iaC

ypru

sLa

tvia

Hun

gary

Lith

uani

aS

lova

kia

Pola

ndBu

lgar

iaM

alta

NM

S 1

2 [a

]

EU

27

[a]

Cro

atia

Turk

eyM

KD

[b]

Eur

o 4

coun

tries

[a]

Nor

way

Pre

vale

nce

(%)

Difficulty paying bills most of the time No difficulty paying bills

Males

Source: data from EQLS, 2007.Notes: [a] average of national rates; [b] MKD: ISO code for the former Yugoslav Republic of Macedonia.

Looking at the impact of employment on the lack of access to recreational or green areas, the data indicate that its impact is stronger but less consistent for males, and weaker but more consistent in females. The reported lack of access is consistently slightly higher among unemployed than employed females in EU15 countries (33% versus 31%), NMS12 countries (46% versus 44%) and Euro 4 countries (57% versus 55%). Among males the disadvantage is only higher among unemployed than employed


males in EU15 countries (36% versus 30%); in contrast, it is lower among unemployed than employed males in NMS12 countries (38% versus 41%) and Euro 4 countries (41% versus 48%). Across all EU27 countries, the prevalence of complaints about lack of access to recreational or green areas shows little variation by employment: the prevalence reported by females is 39% (unemployed) and 37% (employed), while in male respondents it is 37% (unemployed) and 35% (employed).9

At the aggregated level, there are reversed social gradients showing an increasing lack of access with a higher level of education in all subregions and for both females and males (see Table 10). The gradient is most obvious in the Euro 4 countries and much less clearly expressed in the EU15 countries. In all subregions, differences by education level are much more pronounced than differences by sex.

However, in several individual countries the prevalence of complaints about lack of access to recreational or green areas is much higher among the less-educated than the better-educated. This is the case in Finland (17% versus 6%), the former Yugoslav Republic of Macedonia (63% versus 46%), Ireland (50% versus 20%), Italy (81% versus 66%), Malta (67% versus 50%) and the Netherlands (40% versus 19%) for females, and in Finland (20% versus 5%), France (43% versus 32%), Germany (50% versus 14%), Malta (60% versus 53%) and the United Kingdom (28% versus 14%) for males.

Finally, the impact of household composition on lack of access to recreational or green areas was considered. The lack of access is more often expressed in single-parent households than in households of couples without children in all subregions. A considerably higher proportion of single-parent households are female than male (EU15: 9% female versus 2% male, NMS12: 12% versus 3%, Euro 4: 9% versus 2%; prevalence calculated among all EQLS participants of the respective subregion). Female single-parent households have higher prevalence rates of complaints about lack of access to recreational or green areas compared to couples with children in EU15 and NMS12.

An analysis of the impact of sex on inequality among single-parent households revealed that a distinct difference (with a higher lack of access among female than male single-parent households) was seen in EU15 countries (36% of female, 26% of male single-parent households) and, less strongly, in Euro 4 countries (49% of female, 45% of male single-parent households), but not in NMS12 countries (41% of female, 42% of male single-parent households).10

Beyond the impact of social and demographic determinants, the Spanish fact sheet on lack of access to recreational or green areas (see Annex 1) shows that geography and climate conditions can also affect the provision of such spaces to the public.

Target groups for actionThe impact of all the respective socioeconomic or demographic determinants stratified by sex was summarized (see Fig. 38), showing diverse and partially conflicting trends by subregion and sex. The population groups with a higher prevalence of complaints about lack of access to recreational or green areas are:• males and females in EU15, males in NMS12 and females in Euro 4 countries with difficulty paying

bills most of the time;• those in the lowest income quartile in EU15, and those in the highest income quartile in NMS12

countries;• unemployed females in all regions and unemployed males in EU15, as well as employed males in

Euro 4 countries;• all those with a higher education level in NMS12 and Euro 4 countries (in EU15 this applies

predominantly to males);• female single-parent households in all regions (with the strongest effect visible in EU15) and male

single-parent households in NMS12 and Euro 4 countries.

9 Data for EU15, NMS12, EU27 and Euro 4 countries represent averages of national rates�10 Data for EU15, NMS12, EU27 and Euro 4 countries represent averages of national rates�


Table 10. lack of access to recreational or green areas by education level and sex (2007)

Female male

low education

medium education

High education

low education

medium education

High education


38�0 47�9 58�0 33�7 44�0 52�6

Source: data from EQLS, 2007.Note: [a] average of national rates.

Health implications Although no quantitative risk assessment of the health effects of characteristics of the built environment such as access to recreational or green areas has been performed thus far, the importance of design of the built environment and access to the natural environment for physical activity and active living has repeatedly been emphasized (Edwards and Tsouros, 2006). Green space is understood as a resource for physical and mental health as well as for well-being (Maas et al., 2006; Mitchell and Popham, 2008; Abraham, Sommerhalder and Abel, 2010).

It is very likely that the observed social differences in complaints about lack of access to recreational or green areas have an impact on social inequalities in physical activity and health.

Conclusions and suggestionsThe frequency of self-reported reasons to complain about lack of access to recreational or green areas varies widely between countries, so the following conclusions, which are based on subregional averages, may not be true for each individual country.

Social inequalities in access to recreational or green spaces exist, indicated by lower reported access levels among socially disadvantaged individuals, such as those characterized by the single-parent household type indicator in all subregions, and those with difficulty paying bills most of the time, primarily in EU15 countries (see Fig. 38). Unemployed females consistently report a slightly higher lack of access in all regions, as well as unemployed males in EU15, and both females and males in the lowest income quartile in EU15.

On the other hand, social disadvantage is not by default an indicator of lower levels of access to recreational or green areas; individuals with a high education level in all regions report a lack of access more often, as well as employed males in NMS12 and Euro 4 countries and individuals with an income in the highest quartile in NMS12 countries, for example. Thus, the direction and degree of social inequalities in lack of access to recreational or green areas depends on the local or regional situation in a given country or region, as well as on the socioeconomic indicator analysed. However, the seemingly logical expectation that disadvantaged groups might be more affected by a lack of access to recreational or green areas is mostly met in the EU15 region, while NMS12 and Euro 4 countries show more variation.

Combined exposures could be analysed using additional data from the EQLS. For example, question 54 of the EQLS covers reasons to complain about noise, air pollution, water quality, crime, violence or vandalism, and litter or rubbish in the streets in the immediate neighbourhood of the respondent’s home. Looking at more aspects of the built environment could help to build an understanding of differences in inequality by sex, because safety issues are often more relevant for females.

Local governments play a crucial role in creating healthy built environments. The cooperation of several sectors – such as urban planning, housing, transport and public health sectors – is required (Edwards and Tsouros, 2006).


Fig. 38. Prevalence of complaints about lack of access to recreational or green areas by socioeconomic indicator and sex (2007)

Females

0 10 20 30 40 50 60

Difficultypaying bills

Income

Employment

Householdtype

Level of education

Males

0 10 20 30 40 50 60

Difficulty paying bills

Income

Employment

Householdtype

Level of education

EU15: high social position EU15: low social positionNMS12: high social position NMS12: low social positionEuro 4: high social position Euro 4: low social position

Source: data from EQLS, 2007.Note: low social position represents low level of education, single-parent household, unemployed, lowest income quartile and difficulty paying bills most of the time, respectively; high social position represents high level of education, couple without children, employed, highest income quartile and no difficulty paying bills, respectively.

Suggested mitigation actions are:• creation and maintenance of healthy built environments with accessible recreational and green

areas of good quality;• gaining the cooperation of urban planning, housing, transport and public health sectors in

health inequality impact assessments;• better reporting on the quality, dimension, ease and convenience of access, and perception of

recreational and green areas by sex and socioeconomic group; • better research into combined exposures to hazards and into resources beneficial to health in

the built environment, including their distribution by sex and social position.


INEquAlITIES IN SECONd-HANd SmOKE ExPOSurE AT HOmE

Introduction Second-hand smoke is one of the most important and widespread exposures in the indoor environment (Öberg et al., 2010). The harmful health effects of second-hand smoke have been intensively evaluated (U.S. Department of Health and Human Services, 2006; IARC, 2004). Social disparities in exposure to second-hand smoke at home in terms of a higher exposure in socially disadvantaged groups are well documented in Europe, with many studies focusing on children (Bolte and Kohlhuber, 2005; Bolte and Fromme, 2009; see also the fact sheet on potential smoke exposure at home by SES from Germany in Annex 1).

Indicator analysis: inequalities by sex, self-assessed social position, difficulty paying bills and employmentAll prevalence figures quoted are based on analysis restricted to non-smokers. Prevalence of potential second-hand smoke exposure at home varies between 4% and 61% by country (4–62% among males; 4–65% among females). Across EU27, 24% of male and 25% of female respondents report potential exposure to second-hand smoke at home. Prevalence is 25% for both males and females in EU15 countries and 24% for both sexes in NMS12 countries. There are much higher prevalence rates in the three Euro 4 countries (49% of males, 57% of females).

Differences by sex exist in most countries, but in different directions: potential exposure is higher in males than females in 13 countries and higher in females than males in 13 countries. In four countries there is no difference by sex. There is no consistent pattern for age groups across the countries.

Self-assessed position on the social scale is associated with potential exposure to second-hand smoke at home in most countries (see Fig. 39). In the majority of countries, those indicating a low position on the social scale report potential exposure more often. Across EU27, prevalence rates are 30% for males with low versus 24% for males with high social position, and 30% for females with low versus 23% for females with high social position. In EU15 countries, the prevalence is 31% for both males and females with low social position versus 23% for both sexes with high social position. In NMS12 countries, the reported potential exposure for low versus high social position is 30% versus 25% among males and 28% versus 24% among females; in Euro 4 countries, it is 53% versus 48% among males and 57% versus 51% among females.

Some countries (for example, Finland, Estonia, Hungary and Malta for females and Denmark, Finland, the Netherlands, Cyprus and Estonia for males) show the reverse inequality pattern, with more frequent potential exposure among individuals with high social position.

The overall picture across all subregions is that individuals with difficulty paying bills most of the time are more frequently exposed to potential second-hand smoke at home than those with no difficulty (see Table 11). Moreover, in all regions – except in the case of females in EU15 and Euro 4 countries – there is a consistent social gradient showing an increase in potential smoke exposure at home with increasing difficulty paying bills. The largest inequality among those with difficulty paying bills most of the time versus those with no difficulty is observed for males in Euro 4 countries.

Despite the rather consistent trends, there are large differences between individual countries within each subregion, most strongly expressed among males in EU15 countries (see Fig. 40). Among those with difficulty paying bills most of the time, Euro 4 countries Croatia and the former Yugoslav Republic of Macedonia show the highest prevalence rates for both males and females, but EU15 countries Sweden and Austria also show high prevalence levels for males.


Fig. 39. Prevalence of potential exposure to second-hand smoke at home by self-assessed social position and sex (2009)

Females

0

10

20

30

40

50

60

70

Finl

and

Swed

enP

ortu

gal

Irela

ndG

erm

any

Fran

ceU

nite

d Ki

ngdo

mBe

lgiu

mLu

xem

bour

gA

ustri

aIta

lyN

ethe

rland

sSp

ain

Den

mar

kG

reec

eE

U 1

5 [a

]

Lith

uani

aH

unga

ryC

zech

Rep

ublic

Slov

enia

Mal

taS

lova

kia

Est

onia

Bulg

aria

Rom

ania

Latv

iaPo

land

Cyp

rus

NM

S 1

2 [a

]

EU

27

[a]

Turk

eyC

roat

iaM

KD

[b]

Eur

o 4

coun

tries

[a]

Pre

vale

nce

(%)

Males

0

10

20

30

40

50

60

70

Finl

and

Ger

man

yN

ethe

rland

sU

nite

d Ki

ngdo

mP

ortu

gal

Swed

enFr

ance

Belg

ium

Luxe

mbo

urg

Irela

ndD

enm

ark

Italy

Aus

tria

Spai

nG

reec

eE

U 1

5 [a

]

Slo

vaki

aLi

thua

nia

Est

onia

Cze

ch R

epub

licM

alta

Hun

gary

Latv

iaBu

lgar

iaSl

oven

iaC

ypru

sR

oman

iaPo

land

NM

S 1

2 [a

]

EU

27

[a]

Turk

eyM

KD

[b]

Cro

atia

Eur

o 4

coun

tries

[a]

Pre

vale

nce

(%)

Low social position High social position

Source: data from Special Eurobarometer 332: tobacco; data analysis restricted to non-smokers.Notes: [a] average of national rates; [b] MKD: ISO code for the former Yugoslav Republic of Macedonia.

Table 11. Prevalence of potential exposure to second-hand smoke at home by difficulty paying bills and sex (2009)

Female male

No difficulty



No difficulty



EU15 [a] 23�3 26�7 26�5 23�9 26�3 30�7NMS12 [a] 25�3 26�6 34�7 25�1 27�1 30�2EU27 [a] 24�2 26�7 30�2 24�4 26�6 30�5Euro 4 [a] 53�1 62�2 58�4 46�6 51�1 62�2

Source: data from Special Eurobarometer 332: tobacco (TNS Opinion and Social, 2010); data analysis restricted to non-smokers.Note: [a] average of national rates.

Among males, there is distinct variation in levels of potential exposure between employed and unemployed individuals across countries. In 17 of the 30 countries the prevalence of potential exposure to second-hand smoke at home is higher among unemployed than employed males. In females, on the other hand, the prevalence differences are more consistent, showing a higher exposure rate among unemployed than employed females in 26 of the 30 countries.


Across EU27, 29% of unemployed versus 23% of employed males and 32% of unemployed versus 23% of employed females are potentially exposed to second-hand smoke at home. Exposure prevalence in unemployed and employed individuals differs less in EU15 countries (males: 23% versus 21%, females: 27% versus 21%) than in NMS12 countries (males: 36% versus 24%, females: 38% versus 25%), while in Euro 4 countries a large difference by employment status is found for females but not for males (males: 48% versus 47%, females: 67% versus 47%).11

Fig. 40. Prevalence of potential exposure to second-hand smoke at home by difficulty paying bills and sex (2009)

Females

0102030405060708090

Den

mar

k [a

]Fi

nlan

d [a

]Sw

eden

[a]

Luxe

mbo

urg

Ger

man

yP

ortu

gal

Fran

ceIre

land

Uni

ted

King

dom

Aus

tria

Spai

nIta

lyG

reec

eN

ethe

rland

sBe

lgiu

mE

U 1

5 [b

]

Lith

uani

aH

unga

ryC

zech

Rep

ublic

Est

onia

Slov

enia

Mal

taR

oman

iaBu

lgar

iaLa

tvia

Pola

ndS

lova

kia

Cyp

rus

NM

S 1

2 [b

]

EU

27

[b]

Turk

eyC

roat

iaM

KD

[c]

Eur

o 4

coun

tries

[b]

Pre

vale

nce

(%)

Males

0102030405060708090

Den

mar

k [a

]Fi

nlan

d [a

]U

nite

d K

ingd

om [a

]N

ethe

rland

sIta

lyFr

ance

Irela

ndP

ortu

gal

Belg

ium

Gre

ece

Ger

man

ySp

ain

Luxe

mbo

urg

Swed

enA

ustri

aE

U 1

5 [b

]

Mal

taC

zech

Rep

ublic

Est

onia

Lith

uani

aSl

oven

iaR

oman

iaH

unga

ryPo

land

Latv

iaBu

lgar

iaS

lova

kia

Cyp

rus

NM

S 1

2 [b

]

EU

27

[b]

Turk

eyM

KD

[c]

Cro

atia

Eur

o 4

coun

tries

[b]

Pre

vale

nce

(%)

Difficulty paying bills most of the time No difficulty paying bills

Source: data from Special Eurobarometer 332: tobacco (TNS Opinion and Social, 2010); data analysis restricted to non-smokers.Notes: [a] prevalence by difficulty paying bills most of the time = 0% (due to low numbers in the respective social group); [b] average of national rates; [c] MKD: ISO code for the former Yugoslav Republic of Macedonia.

Target groups for actionThe impact of all the respective socioeconomic determinants stratified by sex was summarized (see Fig. 41). There is a consistent pattern of socioeconomic inequalities in second-hand smoke exposure at home, irrespective of the socioeconomic indicator used. Higher exposure prevalence is present among individuals with low self-assessed social position, among those with difficulty paying bills most of the time and among unemployed individuals (mostly females, and especially in Euro 4 countries). Main target groups for action regarding smoking in the home are thus groups such as unemployed females and males with financial difficulties.

11 Data for EU15, NMS12, EU27 and Euro 4 countries represent averages of national rates�


Fig. 41. Prevalence of potential exposure to second-hand smoke at home by socioeconomic indicator and sexFemales

0 10 20 30 40 50 60 70

Employment


Socialposition

Males

0 10 20 30 40 50 60 70

Employment


Socialposition

Prevalence (%)EU15: high social position EU15: low social positionNMS12: high social position NMS12: low social positionEuro 4: high social position Euro 4: low social position

Source: data from Special Eurobarometer 332: tobacco (TNS Opinion and Social, 2010); data analysis restricted to non-smokers.Note: low social position represents low self-assessed position on the social scale, difficulty paying bills most of the time and unemployed, respectively; high social position represents high self-assessed position on the social scale, (almost) no difficulty paying bills, and employed, respectively.

Health implicationsWorldwide exposure to second-hand smoke and its burden of disease in children and adult non-smokers has been estimated for the year 2004 (Öberg et al., 2011). About 1.0% of worldwide mortality and 0.7% of total worldwide burden of disease are attributable to the exposure of non-smokers to second-hand smoke, the largest disease burdens including 5.9 million DALYs lost from lower respiratory tract infections in children younger than 5 years, 2.8 million DALYs lost from ischaemic heart disease in adults, and 1.2 million DALYs lost from asthma in adults. Globally, 47% of deaths from second-hand smoke occurred in women, 28% in children, and 26% in men.

The higher potential exposure to second-hand smoke at home among the socially disadvantaged, as indicated by the Special Eurobarometer 332: tobacco data, is likely to contribute to health inequalities.

Conclusions and suggestionsThere is wide variation in the prevalence of self-reported potential exposure to second-hand smoke at home between countries. The Special Eurobarometer 332: tobacco report summarizes that the level of smoking permissiveness in private settings is correlated with the proportion of smokers in a given country (TNS Opinion and Social, 2010).

Social inequalities in second-hand smoke exposure at home exist with higher exposure among socially disadvantaged groups, characterized by low self-assessed social position, difficulty paying bills most of the time and unemployment (see Fig. 41).


Legislation creating smoke-free public places is mandated in Article 8 of the WHO Framework Convention on Tobacco Control, in order to protect people from the harmful consequences of second-hand smoke exposure at the population level (WHO, 2003). Population-based tobacco control measures might reduce population smoking prevalence, and might therefore also reduce second-hand smoke exposure.

Smoking bans are one component of a comprehensive tobacco control programme. However, according to a recent Cochrane review, legislation-based smoking bans diminish second-hand smoke exposure in public places but do not change self-reported second-hand smoke exposure at home (Callinan et al., 2010). Policy and prevention measures providing education and raising awareness about the health hazards of second-hand smoke exposure focusing on individual level intervention have been shown to be less effective in socially disadvantaged groups. Thus, greater efforts need to be made to target interventions specifically to reach socially disadvantaged people.

Suggested mitigation actions are:• creation and maintenance of smoke-free public places by full-scale implementation of the

WHO Framework Convention on Tobacco Control;• targeted interventions aiming to diminish second-hand smoke exposure at home, particularly

addressing socially disadvantaged groups;• better reporting on second-hand smoke exposure at home by sex and socioeconomic group,

including possible effects of population-based tobacco control measures.

INEquAlITIES IN SECONd-HANd SmOKE ExPOSurE AT wOrK

IntroductionThe health effects of second-hand smoke exposure are well known and have been intensively reviewed (U.S. Department of Health and Human Services, 2006; IARC, 2004; Öberg et al., 2011). Comparable to the social disparities shown in exposure to second-hand smoke at home, individuals in lower socioeconomic groups may experience a higher risk of second-hand smoke exposure at work (Moussa, Lindström and Ostergren, 2004). A study in Germany has shown that employees with a higher education level are more likely to work in smoke-free workplaces or workplaces with at least partial smoking restrictions (Ruge et al., 2010).

Indicator analysis: inequalities by sex, self-assessed social position, difficulty paying bills and occupationAll prevalence figures quoted are based on analysis restricted to non-smokers. Prevalence of exposure to tobacco smoke indoors at work varies between 4% and 46% by country (males: 3–54%, females: 3–40%). Prevalence rates are 23% of males and 15% of females across EU27, 19% of males and 13% of females in EU15 countries, 27% of males and 18% of females in NMS12 countries, and 25% of males and 22% of females in the three Euro 4 countries covered. Similarly, in 24 of the 30 individual countries, prevalence levels are higher among males than females, indicating the general existence of differences by sex in occupational second-hand smoke exposure.

Across all countries, there is considerable variation in exposure by self-assessed position on the social scale, which is especially expressed among males (see Fig. 42). Across EU27, 28% of males with low social position are exposed to second-hand smoke at work in contrast to 20% of males with high social


position. There is no such prevalence difference for females. Inverse social gradients with a higher exposure among individuals with a low social position are observed for both sexes in EU15 countries but only for males in NMS12 countries, where the opposite trend is found for females. In Euro 4 countries there is no difference in prevalence for females, while a higher exposure level is reported for males with high self-assessed social position.

Fig. 42. Prevalence of exposure to second-hand smoke at work by self-assessed social position and sex (2009)

Females

0

10

20

30

40

50

60

70

Den

mar

k [a

]Ire

land

[a]

Swed

en [a

]U

nite

d K

ingd

om [a

]G

erm

any

Finl

and

Luxe

mbo

urg

[b]

Belg

ium

Fran

ceP

ortu

gal

Aus

tria

Net

herla

nds

Italy

Gre

ece

Spai

nE

U 1

5 [c

]

Mal

ta [a

]La

tvia

Slov

enia

Est

onia

Lith

uani

aC

zech

Rep

ublic

Slo

vaki

aC

ypru

sR

oman

iaH

unga

ryBu

lgar

iaPo

land

NM

S 1

2 [c

]

EU

27

[c]

Turk

eyC

roat

iaM

KD

[d]

Eur

o 4

coun

tries

[c]

Pre

vale

nce

(%)

Males

0

10

20

30

40

50

60

70

Swed

en [a

]U

nite

d K

ingd

om [a

]P

ortu

gal

Irela

ndD

enm

ark

Fran

ceG

erm

any

Net

herla

nds

Finl

and

Belg

ium

Italy

Aus

tria

Spai

nLu

xem

bour

gG

reec

eE

U 1

5 [c

]

Rom

ania

Slov

enia

Latv

iaLi

thua

nia

Cze

ch R

epub

licPo

land

Est

onia

Slo

vaki

aBu

lgar

iaH

unga

ryM

alta

Cyp

rus

NM

S 1

2 [c

]

EU

27

[c]

Turk

eyM

KD

[d]

Cro

atia

Eur

o 4

coun

tries

[c]

Pre

vale

nce

(%)

Low social position High social position

Source: data from Special Eurobarometer 332: tobacco (TNS Opinion and Social, 2010); data analysis restricted to non-smokers.Notes: [a] prevalence by low social position = 0% (due to low numbers in the respective social group); [b] prevalence by high social position = 0% (due to low numbers in the respective social group); [c] average of national rates; [d] MKD: ISO code for the former Yugoslav Republic of Macedonia.

Country data indicate large differences, with most countries showing similar inequality trends for males and females, but some countries – including Belgium, Spain, Cyprus, Estonia and Hungary – present opposing inequalities by social position for males and females. The largest inequalities of double the exposure prevalence are mostly suffered by those with a low self-assessed social position, but doubled prevalence levels can also be identified for individuals with a high social position (for example, in Cyprus and Hungary for females, Romania and Turkey for males).

Compared to the data on prevalence by self-assessed social position, the social differences in second-hand smoke exposure at work by categories of difficulty paying bills vary considerably between countries. In EU27, there are again inverse social gradients, with the highest exposure rate among those with difficulty paying bills most of the time (27% of males, 19% of females), followed by those with difficulty paying bills from time to time (24% of males, 16% of females), and those with no difficulty paying bills (22% of


males, 14% of females). However, it is clear that in EU15 countries, inequalities by difficulty paying bills are stronger for females (prevalence rates ranging from 12% to 19%) while in NMS12 countries, males are more affected (26% to 33%). Compared to the EU countries, the three countries from the Euro 4 subregion show a rather different situation, with no clear trend for females and a reverse trend for males, more frequent exposure being associated with no difficulty paying bills (see Table 12).

Table 12. Prevalence (%) of exposure to second-hand smoke at work by difficulty paying bills and sex, 2009

Female male

No difficulty



No difficulty




25�7 23�2 26�2 29�0 24�0 18�4

Source: data from Special Eurobarometer 332: tobacco (TNS Opinion and Social, 2010); data analysis restricted to non-smokers.Note: [a] average of national rates.

Fig. 43. Prevalence of exposure to second-hand smoke at work by occupation and sex (2009)

Females

0

10

20

30

40

50

60

70

80

Irela

nd [a

]G

erm

any

Finl

and

[a]

Luxe

mbo

urg

[a]

Uni

ted

King

dom

Por

tuga

lSw

eden

[a]

Fran

ceB

elgi

um [a

]D

enm

ark

[a]

Net

herla

nds

[a]

Spai

nIta

ly [a

]G

reec

eA

ustri

aE

U 1

5 [b

]

Latv

iaSl

oven

iaC

zech

Rep

ublic

Slo

vaki

aLi

thua

nia

Mal

taE

ston

iaH

unga

ryR

oman

iaPo

land

Bulg

aria

Cyp

rus

NM

S 1

2 [b

]

EU

27

[b]

Turk

eyC

roat

iaM

KD

[c]

Eur

o 4

coun

tries

[b]

Pre

vale

nce

(%)

Males

0

10

20

30

40

50

60

70

80

Swed

enU

nite

d Ki

ngdo

mP

ortu

gal

Den

mar

k [a

]N

ethe

rland

sIre

land

Ger

man

yFi

nlan

dIta

lyFr

ance

Belg

ium

Gre

ece

Luxe

mbo

urg

[a]

Aus

tria

Spai

nE

U 1

5 [b

]

Slov

enia

Pola

ndS

lova

kia

Cze

ch R

epub

licM

alta

Lith

uani

aLa

tvia

Bulg

aria

Est

onia

Rom

ania

Hun

gary

Cyp

rus

[a]

NM

S 1

2 [b

]

EU

27

[b]

Turk

eyM

KD

[c]

Cro

atia

Eur

o 4

coun

tries

[b]

Pre

vale

nce

(%)

Manual worker Manager

Source: data from Special Eurobarometer 332: tobacco (TNS Opinion and Social, 2010); data analysis restricted to non-smokers.Notes: [a] prevalence among managers = 0%; [b] average of national rates; [c] MKD: ISO code for the former Yugoslav Republic of Macedonia.


When the data are stratified by occupation, a consistent pattern is visible, with few exceptions for either males or females: prevalence of second-hand smoke exposure at work is predominantly higher among manual workers than managers (see Fig. 43). In EU15, inequalities exist for both sexes, with prevalence for males at 23% among manual workers and 13% among managers, while the prevalence for females is 15% among manual workers and 6% among managers. In NMS12 countries, the inequality is less strong for females (20% among manual workers versus 16% among managers) than for males (36% versus 19%). In Euro 4 countries, the same disadvantage is found for males (31% among manual workers versus 13% among managers) but the reverse trend is shown for females, with managers more frequently exposed to second-hand smoke (23% versus 29%).

The national fact sheet on second-hand smoke exposure at work gives further details on the issue in Italy (see Annex 1).

Sensitivity analysisAs described in the data and methods section, exposure to second-hand smoke at work was defined as any exposure, including the category “Less than 1 hour a day”. To assess whether the inclusion of this lowest exposure category may have caused artificial results, the prevalence of second-hand smoke exposure at work was also calculated for only those individuals indicating an exposure of one hour or more per day. Within this classification, 9% of males and 5% of females are exposed to smoke indoors at the workplace across EU27, in EU15 countries it is 7% of males, 5% of females, in NMS12 countries 11% of males, 6% of females, and in Euro 4 countries 12% of males and 9% of females.

Using this more stringent exposure definition, social differences in exposure prevalence remain. Across EU27, exposure prevalence for males is 13.5% among manual workers and 4.8% among managers, and for females is 6.5% among manual workers and 3.2% among managers. This difference persists in EU15 countries (manual workers versus managers: males 10.8% versus 4.3%, females 6.7% versus 1.3%) as well as in NMS12 countries (males 17.0% versus 5.4%, females 6.3% versus 5.7%) and Euro 4 countries (males 13.6% versus 11.1%, females 14.3% versus 4.9%).

Target groups for actionThe impact of all the respective socioeconomic determinants stratified by sex was summarized (see Fig. 44). Increased exposure to second-hand smoke indoors at work shows a consistent pattern of socioeconomic disadvantage for several of the indicators studied. The strongest and most consistent inequality is found for male manual workers, who can therefore be identified as the most important target group for action on occupational exposure. Increased exposure is also found among female manual workers, but it is less strong and does not apply as consistently.

Analysing the impact of self-assessed social position as well as level of difficulty paying bills, a disadvantage can be seen for less resourced individuals in EU countries, while for the Euro 4 countries the focus of action may rather be targeted at the occupational conditions of people – especially males – without financial problems and with high social position.

Health implicationsSecond-hand smoke in the workplace is an important source of exposure for non-smoking adults, especially if they are not exposed at home (U.S. Department of Health and Human Services, 2006). Occupational exposure to second-hand smoke significantly increases the risk of lung cancer and ischaemic heart disease (Stayner et al., 2007; Öberg et al., 2010). More frequent exposure to second-hand smoke indoors at work among socially disadvantaged individuals may contribute to health inequalities.


Several studies have investigated the immediate effect of smoking bans on the health of non-smoking employees in bars and restaurants, and have demonstrated improvements of sensory and respiratory symptoms (see, for example, Menzies et al., 2006; Goodman et al., 2007; Ayres et al., 2009).

Fig. 44. Prevalence of exposure to second-hand smoke at work by socioeconomic indicator and sex (2009)

Females

0 5 10 15 20 25 30 35 40

Occupation


Socialposition

Males

0 5 10 15 20 25 30 35 40

Occupation


Socialposition

Prevalence (%)EU15: high social position EU15: low social positionNMS12: high social position NMS12: low social positionEuro 4: high social position Euro 4: low social position

Source: data from Special Eurobarometer 332: tobacco (TNS Opinion and Social, 2010); data analysis restricted to non-smokers.Note: low social position represents low self-assessed position on the social scale, difficulty paying bills most of the time and manual worker, respectively; high social position represents high self-assessed position on the social scale, (almost) no difficulty paying bills, and manager, respectively.

Conclusions and suggestionsExposure to second-hand smoke indoors at work varies between countries, with an overall higher prevalence in males than females, and higher exposure levels in NMS12 and Euro 4 countries than EU15 countries.

Social inequalities in second-hand smoke exposure at work exist across EU27, with higher exposure prevalence among socially disadvantaged individuals, using the social indicators occupation, financial difficulties, and self-assessed social position (see Fig. 44). For the Euro 4 countries analysed, the inequalities are less consistent, sometimes showing lower exposure prevalence in socially deprived population groups.

At the time of the Special Eurobarometer 332: tobacco survey in 2009 (TNS Opinion and Social, 2010), legislation-based workplace smoking bans were already effective in many European countries. If the survey data are valid, stricter enforcement of existing smoking bans is required. That second-hand smoke exposure may still occur despite a workplace-smoking ban has been demonstrated by a study in the Netherlands, showing exposure especially of males and lower educated employees before and after implementation of the smoking ban (Verdonk-Kleinjan et al., 2009).


Suggested mitigation actions are:• further implementation and enforcement of existing legislation-based smoking bans in

workplaces;• creation and maintenance of smoke-free public places by full-scale implementation of the

WHO Framework Convention on Tobacco Control;• targeted interventions aiming to diminish second-hand smoke exposure at work, addressing

socially disadvantaged employees;• better reporting on second-hand smoke exposure at work by sex and socioeconomic groups,

including the possible effects of population-based tobacco control measures and legislation-based smoking bans in workplaces.

CONCluSION ON ENvIrONmENT-rElATEd INEquAlITIESIt has repeatedly been shown, especially in the case of children, that there are socioeconomic differences in exposure to ambient air pollution, noise and second-hand smoke and in lack of access to green space in Europe (Bolte and Kohlhuber, 2005; Bolte, Tamburlini and Kohlhuber, 2010). This analysis of data from EU-SILC, EQLS and Special Eurobarometer 332: tobacco adds results for adults from up to 30 countries in the WHO European Region to the body of evidence.

In line with the international debate on environmental inequalities and environmental justice, this chapter focuses on social indicators of material circumstances. Whenever possible, stratified analyses were performed to identify the impact of sex on environmental inequalities. Analysis of any modification of the health effects of environmental burdens by socioeconomic factors was not an objective of this overview.

Despite restrictions in data access and applicability of the available data, this systematic approach analysing social differences in environmental exposures shows socioeconomic inequalities in all environmental dimensions studied: noise exposure (or rather annoyance), lack of access to recreational or green areas, and exposure to second-hand smoke at home or at work. Financial limitations, indicated by a low income or self-reported difficulty paying bills, for example, are consistently associated with higher environmental threats, with some exceptions in NMS12 and Euro 4 countries. Other social or demographic determinants, such as household type, also indicate inequalities to the disadvantage of more vulnerable households.

On the other hand, analysis of data stratified by education level, and to some extent also by employment and occupation, shows that in some countries increased environmental exposures may also be associated with higher social status indicated by high education level, employment or managerial positions. Using the available data, there was no way to identify the impact of environmental health awareness (especially in relation to education) and level of expectation on these results.

Since all analyses were based on self-reported data, it would be desirable to strengthen the evidence base with more objective exposure data, as well as data on characteristics of the built environment. Ambient air pollution data are particularly relevant since the data currently available do not allow the assessment of inequalities related to air pollution. Surveys designed with suitable indicators (for example, considering only traffic-related noise) and more precise exposure assessments (such as evidence of real second-hand smoke exposure at home instead of permission to smoke at home) should be implemented. The complexity of the issue cannot be captured by the indicators currently available, so efforts are needed to assess combined or cumulative environmental exposures.


Several types of policy action may be recommended, based on these results: • integration of an inequality perspective into urban planning (for example, in the design of buildings,

streets and activity-friendly neighbourhoods, and in strengthening public transport);• enforcement of existing legislation (such as smoking bans in workplaces);• integration of an inequality perspective into policy measures aimed at reducing environmental

hazards;• targeted measures to protect the most vulnerable groups (individuals with the highest prevalence of

exposure or who are most sensitive to the health effects of environmental threats).

Against the background of a widespread tendency to individualise health, it is essential to embed environmental concerns into the debate on health inequalities for policy-makers. According to the final report of the CSDH, one principle of action is to improve the structural determinants and conditions of daily life; hence, policies must embrace all key sectors of society, not just the health or environment sector (CSDH, 2008).

rEFErENCES

Abraham A, Sommerhalder K, Abel T (2010). Landscape and well-being: a scoping study on the health-promoting impact of outdoor environments. International Journal of Public Health, 55(1):59–69.

Ayres JG, et al. (2009). Bar workers’ health and environmental tobacco smoke exposure (BHETSE): symptomatic improvement in bar staff following smoke-free legislation in Scotland. Occupational and Environmental Medicine, 66(5): 339–346.

Bolte G, Fromme H for the GME Study Group (2009). Socioeconomic determinants of children’s environmental tobacco smoke exposure and family’s home smoking policy. European Journal of Public Health, 19(1):52–58.

Bolte G, Kohlhuber M (2005). PINCHE Project, final report workpackage 5: socioeconomic factors. Gelderland Midden, Public Health Services (http://www.pinche.hvdgm.nl/resource/pdf/documents/final/PINCHE_WP5_final_181105.pdf, accessed 31 May 2011).

Bolte G, Pauli A, Hornberg C (2011). Environmental justice: social disparities in environmental exposures and health. Overview. In: Nriagu JO (ed.) Encyclopedia of Environmental Health. Vol. 2, pp. 459–470. Burlington, Elsevier.

Bolte G, Tamburlini G, Kohlhuber M (2010). Environmental inequalities among children in Europe – evaluation of scientific evidence and policy implications. European Journal of Public Health, 20(1):14–20.

Braubach M, Fairburn J (2010). Social inequities in environmental risks associated with housing and residential location – a review of evidence. European Journal of Public Health, 20(1):36–42.

Brulle RJ, Pellow DN (2006). Environmental justice: human health and environmental inequalities. Annual Review of Public Health, 27:103–124.

Callinan JE et al. (2010). Legislative smoking bans for reducing secondhand smoke exposure, smoking prevalence and tobacco consumption. Cochrane Database of Systematic Reviews, 4:CD005992.

CSDH (2008). Closing the gap in a generation: health equity through action on the social determinants of health. Final report of the Commission on Social Determinants of Health. Geneva, World Health Organization (http://whqlibdoc.who.int/publications/2008/9789241563703_eng.pdf, accessed 7 July 2011).

Directive 2002/49/EC of the European Parliament and of the Council of 25 June 2002 relating to the assessment and management of environmental noise (2002). Brussels, European Commission (http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:32002L0049:EN:HTML, accessed 4 January 2012).

Edwards P, Tsouros A (2006). Promoting physical activity and active living in urban environments: the role of local governments. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0009/98424/E89498.pdf, accessed 29 May 2011).

EQLS [online database] (2007). Essex and Manchester, ESDS (http://www.esds.ac.uk/findingData/eqlsTitles.asp, accessed 28 December 2011).


EU-SILC [online database] (2011). Luxembourg, Eurostat (http://epp.eurostat.ec.europa.eu/cache/ITY_SDDS/EN/ilc_esms.htm#stat_pres, accessed 28 December 2011).

Evans G, Kantrowitz E (2002). Socioeconomic status and health: the potential role of environmental risk exposure. Annual Review of Public Health, 23:303–331.

Gee GC, Payne-Sturges DC (2004). Environmental health disparities: a framework integrating psychosocial and environmental concepts. Environmental Health Perspectives, 112(17):1645–1653 (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1253653/, accessed 18 September 2011).

Goodman P et al. (2007). Effects of the Irish smoking ban on respiratory health of bar workers and air quality in Dublin pubs. American Journal of Respiratory and Critical Care Medicine, 175(8):840–845.

Hänninen O, Knol A (2011). European perspectives on environmental burden of disease: estimates for nine stressors in six European countries. Helsinki, National Institute for Health and Welfare (THL) (http://www.thl.fi/thl-client/pdfs/b75f6999-e7c4-4550-a939-3bccb19e41c1, accessed 26 May 2011).

IARC (2004). Tobacco smoke and involuntary smoking. Lyon, International Agency for Research on Cancer (IARC monographs on the evaluation of carcinogenic risks to humans, Vol. 83; http://monographs.iarc.fr/ENG/Monographs/vol83/mono83.pdf, accessed 5 January 2012).

Kinney PL, O’Neill MS (2006). Environmental equity. In: WHO (ed.), Air quality guidelines: global update 2005. Copenhagen, WHO Regional Office for Europe, pp 135–152 (http://www.euro.who.int/__data/assets/pdf_file/0005/78638/E90038.pdf, accessed 26 May 2011).

Kruize H et al. (2007). Environmental equity and the role of public policy: experiences in the Rijnmond region. Environmental Management, 40(4):578–595.

Laurent O et al. (2007). Effect of socioeconomic status on the relationship between atmospheric pollution and mortality. Journal of Epidemiology and Community Health, 61(8):665–675.

Lee ACK, Maheswaran R (2011). The health benefits of urban green spaces: a review of the evidence. Journal of Public Health, 33(2): 212–222.

Martuzzi M, Mitis F, Forastiere F (2010). Inequalities, inequities, environmental justice in waste management and health. European Journal of Public Health, 20(1):21–26.

Maas J et al. (2006). Green space, urbanity, and health: how strong is the relation? Journal of Epidemiology and Community Health, 60(7):587–592.

Menzies D et al. (2006). Respiratory symptoms, pulmonary function, and markers of inflammation among bar workers before and after a legislative ban on smoking in public places. Journal of the American Medical Association, 296(14):1742–1748.

Mitchell R, Popham F (2008). Effect of exposure to natural environment on health inequalities: an observational population study. Lancet, 372(9650):1655–1660.

Mitchell R, Astell-Burt T, Richardson EA (2011). A comparison of green space indicators for epidemiological research. Journal of Epidemiology and Community Health, 65(10):853–858.

Moussa K, Lindström M, Ostergren PO (2004). Socioeconomic and demographic differences in exposure to environmental tobacco smoke at work: the Scania Public Health Survey 2000. Scandinavian Journal of Public Health, 32(3):194–202.

Öberg M et al. (2010). Second-hand smoke: assessing the burden of disease at national and local levels.Geneva, World Health Organization (Environmental Burden of Disease Series, No. 18; http://www.who.int/quantifying_ehimpacts/publications/SHS.pdf, accessed 29 May 2011).

Öberg M et al. (2011). Worldwide burden of disease from exposure to second-hand smoke: a retrospective analysis of data from 192 countries. Lancet, 377(9760):139–146.

O’Neill MS et al. (2007). Poverty, environment, and health: the role of environmental epidemiology and environmental epidemiologists. Epidemiology, 18(6):664–668.

Pruss-Üstun A, Corvalán C (2006). Preventing disease through healthy environments. Towards an estimate of the environmental burden of disease. Geneva, World Health Organization (http://www.who.int/quantifying_ehimpacts/publications/preventingdisease.pdf, accessed 26 May 2011).


Richardson EA, Mitchell R (2010). Gender differences in relationships between urban green space and health in the United Kingdom. Social Science and Medicine, 71(3):568–575.

Ruge J et al. (2010). Workplace smoking restrictions: smoking behavior and the intention to change among continuing smokers. International Journal of Public Health, 55(6):599–608.

Stayner L et al. (2007). Lung cancer risk and workplace exposure to environmental tobacco smoke. American Journal of Public Health, 97(3):545–551.

TNS Opinion and Social (2010). Special Eurobarometer 332: tobacco. Brussels, European Commission (http://ec.europa.eu/public_opinion/archives/ebs/ebs_332_en.pdf, accessed 31 May 2011).

U.S. Department of Health and Human Services (2006). The health consequences of involuntary exposure to tobacco smoke: a report of the surgeon general. Atlanta, GA, U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, Coordinating Center for Health Promotion, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health (http://www.surgeongeneral.gov/library/secondhandsmoke/, accessed 07 July 2011).

Verdonk-Kleinjan WM et al. (2009). Does the workplace-smoking ban eliminate differences in risk for environmental tobacco smoke exposure at work? Health Policy, 92(2–3):197–202.

WHO (2003). WHO Framework Convention on Tobacco Control. Geneva, World Health Organization (http://www.who.int/tobacco/framework/WHO_FCTC_english.pdf, accessed 17 October 2011).

WHO (2010). Environment and health risks: a review of the influence and effects of social inequalities. Copenhagen, WHO Regional Office for Europe (http://www.euro.who.int/__data/assets/pdf_file/0003/78069/E93670.pdf, accessed 29 May 2011).

WHO (2011). Burden of disease from environmental noise: quantification of healthy life years lost in Europe. Copenhagen, WHO Regional Office for Europe (http://www.who.int/quantifying_ehimpacts/publications/e94888.pdf, accessed 26 May 2011).

5

GAPS IN EVIDENCE AND RESTRICTIONS

ON ASSESSING ENVIRONMENTAL

HEALTH INEQUALITIES

CHAPTER

Chapter 5. Gaps in evidence and restrictions on assessing environmental health inequalities116

CHAPTEr 5. GAPS IN EvIdENCE ANd rESTrICTIONS ON ASSESSING ENvIrONmENTAl HEAlTH INEquAlITIESNita Chaudhuri, Matthias Braubach

INTrOduCTIONHigh-quality data should be the foundation for understanding and taking political action on environmental health inequalities in Europe. However, one of the greatest weaknesses in work on environmental health inequalities is the many gaps in data and evidence in the region. The lack of environmental risk factor and exposure data – as well as the limited stratification of environmental risk exposures by factors such as age, gender, income, ethnicity, employment and education – makes it difficult to identify vulnerable groups and the magnitude of inequality. Inconsistent inequality data also make producing comparisons within and between countries a challenge, and potentially limit the development of targeted policy responses. This can have important consequences for the health of populations, and thus the lack of data on environmental health inequalities in itself represents an environmental justice issue.

Timely, accurate, valid, reliable and relevant data help to identify important associations between environmental risk factors and sociodemographic determinants, and lead to better understanding of the patterns of exposure-response relationships and subsequent health outcomes. Such increased understanding and awareness will inform planning and decision-making for interventions and resource allocation, and can thus help to reduce the disproportionate burden of environmental risks on vulnerable groups.

Improving the availability and quality of such data is a major priority for further and more detailed assessments of environmental health inequalities. In consequence, the current data limitations – described in more detail below – need to be tackled urgently to implement sociodemographic determinants of health approaches effectively within the environmental health sector.

mISSING dATALack of data on environmental exposure is one key reason why a complete picture of regional environmental health inequalities is not possible. This is especially the case for the eastern part of the WHO European Region, where information on environmental risks is scarce in general and thus already restricts environmental health risk assessments on population level. However, even in the western part of the Region, a range of environmental exposure indicators is only sporadically collected in many countries, and for various specific environmental risks no reliable data are available in any of the Member States.

Typical environmental justice indicators – such as distance to waste site, location close to busy road, location in dangerous area (prone to floods, landslides, and so on) – are rarely collected by countries, and very few of these indicators link neighbourhood and housing location to social and/or environmental


deprivation. For other relevant environmental risks considered high priority – such as air pollution, indoor air quality, chemical exposure, radiation or measured noise levels – reliable statistical data are also limited to very few countries.

lImITEd STrATIFICATION by SOCIOdEmOGrAPHIC dETErmINANTSAlthough a variety of environmental risk factor data have been gathered by both national and international bodies (mostly restricted to the EU region), stratification of these risks by sociodemographic determinants is limited.

International databases such as Eurostat, for example, provide a number of environmental risk indicators for many countries in the EU. Several relevant sociodemographic dimensions of these data (such as education, employment status and ethnicity), however, are not available. Stratification of data on important risk factors such as water supply, collected by the Joint Monitoring Programme, is limited to rural–urban dimensions, which may not be relevant to all countries. Injury data provided in the WHO European Health for All database are also limited to stratification by age and sex only, leaving out, for example, the very relevant dimension of income. Environmental databases, while providing rich information on environmental conditions and trends, usually focus exclusively on this data dimension and seldom offer further breakdown of the data by population subgroups.

National statistical data are often affected by similar limitations. For example, data on air pollution are mostly restricted to the assessment of pollutant levels and are not related to sociodemographic population variables. Similarly, limited data on indoor air quality are available at the national level for some countries but cannot be broken down by sociodemographic determinants. This limits the richness of the inequality information that can be reported.

Many countries have collected environmental risk data – on traffic noise, for example – in various formats such as geographical information systems (GIS) or exposure maps. Information on sociodemographic characteristics, however, has not been collected simultaneously, and the potential overlay of mapped data on sociodemographic determinants has yet to be explored in many countries. Aggregate-level data (on districts, postal codes, deprived neighbourhoods, and so on) have also been collected; they indicate useful dimensions of environmental risks associated with social or neighbourhood deprivation levels, but do not enable the identification of environmental health inequalities at the household or personal level. While the assessment of territorial disparities is enhanced by such data collection techniques, the identification of vulnerable risk groups remains restricted.

Social deprivation indices have been used and suggested by a number of countries to identify populations at risk and provide a powerful tool for identifying major risk groups. However, such indices would need to be disaggregated to provide a more detailed understanding of the link between the separate social determinants and environmental risk factors.

dATA quAlITyAccuracy, validity, timeliness, completeness and relevance are some of the desirable attributes that define data quality. Infrastructure and resource availability for data collection, as well as country and regional priorities, often influence the extent to which quality can be assured. In the WHO European Region there is a clear trend that when an increased level of detail or a higher quality of data is required, fewer countries are able to report on it. Also, greater variations and inconsistencies between national data exist when more detailed or objectively measured data are reported.

A key challenge for environmental risk assessment is that some exposure data tend to be based on self-reported rather than objective data, which means that perception rather than objective measurement is the basis for the assessment. Although it is often meaningful to ask individuals about their subjective


perception of exposures, rather than to consider only objective data that people may feel about very differently, self-reported data make comparison difficult exactly because they include a personal assessment. Examples of environmental exposure that are often surveyed using self-reported exposure include noise, dampness and air pollution data. The restricted validity and reliability of these self-reported rather than objective measures need to be considered when assessing survey designs or inequality relationships based on such data. However, for many environmental risk factors such information will be the only data available and, despite its limitations, may still provide useful indications of the perceived environmental burden.

Representativeness of data can also be problematic, especially in relation to data collected through national surveys that are not developed to provide detailed data on specific population subgroups. For example, small risk groups – such as poor single-parent households – may make up only a small percentage of the population, and consequently may be represented by a low number of people responding to the survey, leading to vague and, at best, indicative results. Extrapolation of such data to the national level therefore causes substantial concerns regarding the reliability of national assessments. Nevertheless, acknowledging this limitation, many inequality assessments will have to rely on such surveys since no alternative data sources are available. Targeted surveys on specific risk groups, based on adequate sample sizes, would be necessary for more reliable and detailed assessments.

CONSISTENCy ANd COmPArAbIlITyConsistency is another dimension of data quality that allows for comparability within and between countries. The highest consistency levels are usually provided by international studies using similar methods and definitions for each country surveyed, but these studies tend to be more general and address the full populations. By contrast, national studies often provide more detailed data and/or focus on specific target groups, but tend to be difficult to compare to similar studies undertaken in other countries.

Several criteria influence the extent to which data on environmental risk factors and sociodemographic determinants are consistently collected and reported. Numerous definitions exist, for example, within and between countries on many environmental and sociodemographic indicators, which include both subjective and objective measures. Interpretations of given risk factors may differ between countries because of sociocultural, environmental or health priorities. This leads to the development of different types of measurement indicators. For example, assessments of overcrowding may measure floor space in square metres or number of rooms, or may report on subjective notions of overcrowding. Similarly, units of measurement may also differ, including individuals, households, or percentages of the population. This makes consistency in reporting difficult, especially when individual national studies from different countries are compared and represent the main or only data source.

Similarly, countries may also collect multiple variables for one risk factor where definitions and data collection protocols may be different, again creating inconsistency in comparability. For example, various collection mechanisms exist for data on injuries and on absence from work due to injuries, making country comparison difficult. A similar situation is created by noise exposure studies which use different exposure thresholds, indicating the proportion of households exposed to more than 60 A-weighted decibels (dB(A)) in one country, while identifying households with exposure levels above 65 dB(A) in another.

Indicators can also be collected at different intervals or frequencies, making comparability between countries within specific time frames tricky. This may be the case particularly for detailed national population surveys or censuses that are not implemented on an annual basis, but also relates to international surveys. This aspect of data quality also affects the timeliness of the data to provide reliable information for decision-making to address environmental health inequalities.


Geographic scope also differs, as similarly defined indicators may be collected at local, regional or national levels. This makes it difficult to compare the magnitude of inequalities between and within countries, and is further complicated by the fact that many environmental exposures do not adhere to territorial or administrative boundaries.

Sociodemographic data often vary in relation to country-specific categories. Levels of education, for example, differ according not only to years of obligatory schooling, but also to the type of schooling, such as vocational or university. These distinctions are not always made and the categories used by countries are often not comparable. Similarly, income categories may differ by stratification of actual levels of income in local currency, or based on different definitions of high-, low- or middle-income groups. Similar problems apply for data on ethnic groups or migrants, and some countries even have legislation that bans the identification of survey participants in relation to ethnic groups.

ACCESS TO dATADespite all such limitations, many data useful for the assessment of environmental health inequalities are frequently collected by a range of actors working on environment, health, social protection, urban planning, transport, occupational safety and so on. However, a lack of coordination and data exchange between different national as well as international institutions prevents the effective exploitation and integration of many data sets to provide a fuller picture of environmental health inequalities. Data elements thus remain scattered and in the hands of different actors instead of being brought together.

In many countries, relevant data are not easily accessible because of legal and cost constraints, and this may also be the case for national or governmental institutions. Public authorities, as well as commercial monopolies, often collect data but are reluctant to release them, and data made available are frequently in formats that are difficult to work with.

CumulATIvE ANd mulTIPlE ExPOSurESIndividuals and households may experience many risk factors in parallel, accumulating problems that lead to multiple exposures. This requires a more complex approach to analysis, as several exposure dimensions need to be brought together within the same data set. However, most risk factors are studied, collected and reported separately, leading to an underestimation of multiple exposures when documenting clustered inequalities. It is difficult to develop a consistent approach that can be similarly applied in a number of countries with different data and priorities. Nevertheless, this should not preclude comparisons between countries using simple messages to convey inequalities.

COuNTry PrIOrITIESCountries have different priorities in relation to environmental health inequalities, which directly affect the availability and use of data. These priorities may relate to levels of development and affluence, historical development or general environmental conditions such as climate or health status. For example, rural–urban differences in some countries represent a significant indicator of inequality, while in other countries such information is considered irrelevant. Similarly, the ability to keep the home warm during winter may not be considered relevant in countries where central heating is a basic standard in all dwellings. Water and sanitation data in many eastern European and central Asian countries are considered to be a priority, but they are not considered relevant for – and are thus no longer collected by – some EU states. In consequence, individual countries have no data available on a selected range of environmental and sociodemographic risk factors.


Sociodemographic inequalities in the eastern European and central Asian countries are among the most notable, yet little has been done in these countries to make the study of the effects of sociodemographic factors on environment and health outcomes a priority. This has led to the dearth of data apparent in this report.

dATA GAPS ANd rElEvANCE FOr PublIC HEAlTHAlthough the literature on environmental health inequalities contributes to the stock of available information – particularly when similar patterns are reported frequently – the overall evidence base is limited, mostly due to the lack of statistical evidence and prevalence studies focusing on specific risk groups. Developing robust and relevant environmental health inequality indicators is, therefore, an essential requirement in order to develop a better understanding of the complex way in which environmental risk factors operate in the social environment, and to identify the respective risk groups to be targeted. Data collected with regular frequency can provide time trend information to inform environmental, health and social actors and to monitor policy responses tackling identified environmental health inequalities.

More work is needed, however, to identify and collect information on relevant environmental risk factors. In particular, data on emerging environmental health issues – such as exposure to toxic chemicals, non-ionizing radiation, nanoparticles or indoor air quality – should be collected regularly, and may provide important insights into differential exposures and impacts on vulnerable groups.

Relevant sociodemographic risk factors also need to be identified and collected, according to country priorities, in order to provide a better picture of those vulnerable groups most at risk. In many countries, for example, data on ethnicity and migrant groups as it relates to environmental risk exposure are rarely collected. Setting up agreements on the priority determinants to be considered is most practical at a national or even subnational level, but this then restricts the use of such data for international comparison.

Extending the description of data constraints, there is an important need to assess and quantify more fully the direct impact of environmental inequalities on health inequalities; this would also help to position this topic as a public health challenge. In order to make the case that disparities in environmental exposure cause health inequalities, a third data dimension is necessary: health outcomes. An ideal database for environmental health inequality assessment would therefore need to include information on health outcomes that can be categorized according to the population subgroups suffering from different exposure levels. This would not only enable an assessment of the inequalities in environmental exposure, but would also extend the assessment to a quantification of health inequalities caused by environmental inequalities. However, it appears that there is no large, international database available that would enable such an assessment.

Furthermore, the type and quality of data collected should take into consideration the context in which social determinants affect environmental conditions where people live, work and play, their differential exposure, biological susceptibility and final health outcome. This could be provided by a description of settings and the dynamic interaction of social determinants with several environmental risk factors simultaneously. For example, social class interacts with other sociodemographic factors such as gender, ethnicity and age to influence access to adequate housing, overcrowding or exposure to chemical substances. Internationally coordinated environmental health inequality studies targeting specific groups and not only full populations could therefore be initiated to examine this.


CONCluSIONMuch work needs to be undertaken to fill the data and evidence gaps that currently restrict the assessment of environmental health inequalities in the WHO European Region. A culture of free information, which requires public authorities and the private sector to give access to environmental information in the spirit of the Aarhus Convention (UNECE, 1998), should be encouraged across countries to make high-quality data available.

A detailed review of data sources and harmonization of criteria for data collection between countries would help to improve data availability, comparability and consistency. Furthermore, building capacity and sharing best practices for those countries with infrastructure and resource constraints could help to increase environmental health inequality data collection. This could include the development of surveys and GIS that simultaneously collect individual and territorial data on health, environment exposure and sociodemographic factors, particularly at country and international levels.

An environmental health inequality network could be set up to facilitate all these activities: in each country one competent agency could be mandated to compile and make available data from all sectors. This agency could also conduct independent analysis and assess national patterns. This would contribute to the development of more useful data and tools for reporting on environmental health inequalities in the region, and would also influence countries to act on the results.

In summary, priority steps to be taken towards the improvement of statistical evidence for environmental health inequality assessment would comprise:• ensuring the right to access to available data at no or low cost;• establishing surveys focusing on target groups in addition to full population surveys;• facilitating the collection of reliable data on priority issues in environmental health;• integrating and increasing the use of analysis of social and demographic variables in

environmental surveys;• developing common tools, methods, definitions and criteria for national work on environmental

health inequalities.

rEFErENCE

UNECE (1998). Convention on access to information, public participation in decision-making and access to justice in environmental matters (Aarhus Convention). Geneva, UNECE (http://www.unece.org/fileadmin/DAM/env/pp/documents/cep43e.pdf, accessed 26 January 2012).

6

PRIORITIES FOR ACTION ON

ENVIRONMENTAL HEALTH INEQUALITIES

CHAPTER

Chapter 6. Priorities for action on environmental health inequalities124

CHAPTEr 6. PrIOrITIES FOr ACTION ON ENvIrONmENTAl HEAlTH INEquAlITIES Matthias Braubach, Stefanie Fleischmann

INTrOduCTIONThe report presents a wide range of inequalities in environmental exposure. One of the most obvious differences displayed is the disparity between individual countries and between the subregions. While these differences are relevant and can indicate the potential for improvement in those countries and subregions performing less successfully, they are not good indicators of the distribution of environmental quality within the respective populations of each country or subregion. Therefore, while acknowledging that the absolute level of environmental problems is relevant, the report goes beyond national and subregional averages, attempting to assess the inequalities in the distribution of environmental risks and to describe the exposure within different population subgroups in both absolute and relative terms.

The figures provided throughout the report identify the environmental health inequalities separately for individual countries and subregions. In this chapter, the findings are condensed and considered from the subregional and national perspectives, identifying the greatest environmental health inequalities and thereby also suggesting potential priorities for action to tackle them.

SuGGESTEd SubrEGIONAl PrIOrITIES FOr ACTIONAlthough the analysis of environmental health inequality is strongly affected by the lack of data from non-EU countries for many of the housing and all the environment inequality indicators, the data do enable the identification of some subregional priorities for action. However, it must be emphasized that – due to the wide diversity of national inequalities – only a few subregional trends for selected inequalities can be identified.• EU15 or Euro 1 subregion: the largest injury-related inequalities occur for transport-related injuries.

With housing-related inequalities, EU15 or Euro 1 countries perform rather well in relation to water and sanitary equipment; conversely, in relation to overcrowding, dampness and thermal comfort, the relative inequalities between population subgroups tend to rise and partially exceed the inequalities seen in the NMS12. Environmental inequality results often indicate that in this subregion inequalities for a range of environmental exposures are among the largest of all the subregions with available data.

• NMS12 or Euro 2 subregion: this subregion is significantly disadvantaged regarding many of the housing-related inequality indicators, and a range of gender inequalities is found for the environmental indicators. Injury-related inequalities are largest for RTIs and fatal poisonings.

• Euro 3 subregion: data for Euro 3 countries are only available for the inadequate water supply indicator and for the injury-related inequality indicators (except work injuries). For all of these indicators, greater inequalities than for some of the other subregions are identified, with the worst


results found for inadequate water supply and fatal poisonings. Euro 3 countries should set as a major priority the development of reliable mechanisms to collect and analyse data on inequalities in environmental risks so that better assessments can be produced in the future.

• Euro 4 subregion: data for Euro 4 countries are only available for the inadequate water supply indicator and for the injury-related inequality indicators (excluding work injuries), as well as for the indicators on second-hand smoke exposure and lack of access to recreational and green areas (for Croatia, the former Yugoslav Republic of Macedonia and Turkey only). Although inequalities are identified for all indicators, the main inequality challenges for Euro 4 countries lie within unintentional injury indicators (transport-related injuries and poisonings) as well as the second-hand smoke exposure indicator. As with the Euro 3 subregion, data collection as a baseline for the assessment of environmental health inequalities must be considered a priority challenge.

As these subregional trends reflect the summarized values for the respective countries, they allow an assessment of the likelihood that countries of a given subregion may face inequality problems related to a particular environmental exposure. Subregions with very obvious inequalities inevitably contain a majority of countries where the respective inequality is an issue. However, it must be cautioned that the subregional priorities can by no means automatically indicate priorities for action at a national level.

SuGGESTEd PrIOrITIES FOr NATIONAl ACTIONTo identify potential priorities for national action among the range of environmental health inequalities, two dimensions of inequality are of relevance. First, data on the relative inequality between population subgroups for a given environmental exposure must be considered. However, as this report shows, high relative inequalities can be found in countries where the absolute population prevalence of this exposure is low, and also where prevalence is relatively low for the selected disadvantaged population subgroups, compared to other countries. Therefore, possible priorities for national action need to be identified with an acknowledgement of both the absolute magnitude of the respective environmental exposure (in total terms, such as for the whole population) and the dimension of relative differences between population subgroups.

Following this approach, data on the relative inequalities and inequality ratios were compiled from the report and graded for each country. A percentile approach was used to categorize the countries into four quartiles, the first quartile containing the countries with the lowest relative inequalities and the fourth quartile containing those with the highest relative inequalities. These data were then brought together with data on the absolute prevalence of the respective risk exposure in the total population – these were also categorized into quartiles, the first quartile containing the countries with the lowest absolute prevalence rates, and the fourth quartile containing those with the highest absolute prevalence rates.

Assessing potential priorities for national action: methodologyTable 13 shows how this approach is applied to an example environmental risk factor in relation to poverty. The columns show the inequality ratio quartiles, with quartile 1 (countries with lowest exposure inequalities in relation to poverty) on the left, and quartile 4 (countries with highest exposure inequalities in relation to poverty) on the right. The rows show the absolute exposure prevalence quartiles for the total population, with quartile 1 (countries with lowest absolute prevalence) at the top, and quartile 4 (countries with highest absolute prevalence) at the bottom.

Table 13 indicates that the combination of the absolute and relative dimensions results in an overall country assessment.• Country A performs well on both inequality dimensions, reporting – in relation to all other

countries – both low absolute exposure prevalence and low inequality by poverty.


Table 13. Potential priorities for national action assessment ranking scheme

quartiles for absolute exposure

prevalence (total population)

quartiles for inequality ratios in exposure prevalence (population below/above poverty threshold)

1 (lowest) 2 3 4 (highest)

1 (lowest) Country A Country b

2 Country G

3 Country H Country d

4 (highest) Country C Country I Country E Country F

Colour coding: Countries with no suggested priority for national action Countries that would be classified as having a suggested priority for national action if they fell into a higher quartile for just one of the two inequality dimensionsCountries with suggested priority for national action (defined as countries being in the fourth quartile for both the absolute and relative inequality dimensions, or in the third quartile for one and in the fourth quartile for the other inequality dimension)

• Country B is in the quartile of countries showing the highest inequality in exposure prevalence between those above and below the poverty threshold, but the absolute population prevalence of exposure to this risk factor is very low. Although the respective inequality needs to be addressed by focusing local action on the population groups that experience disproportionate exposure levels, it can hardly be considered a national priority issue in Country B.

• The opposite case is represented by Country C, which falls into the highest quartile regarding absolute exposure but shows – compared to the other countries – rather modest poverty-related inequality. The respective environmental health risk must therefore be considered a national health challenge in Country C and calls for general public health action. However, as the exposure risk affects those above the poverty threshold almost as much as those below, there is little to gain from targeted interventions aiming at the reduction of inequalities in Country C.

• Finally, some countries combine both disadvantages. Country D is in the third quartile for absolute prevalence and in the fourth quartile with one of the highest relative inequalities by poverty, while Country E shows one of the highest absolute prevalence rates of all countries covered, and also falls into the third quartile of relative inequalities. Country F is categorized in the highest quartile for both dimensions. For countries D, E and F, the respective environmental health risk is greater than in most other countries, and its distribution within the population is also much more unequal than in most other reporting countries. In consequence, they can be identified as countries in which the respective inequality in environmental exposure is – compared to the other countries – expressed most strongly and thus might represent a priority for national action and follow-up.

• In addition to the combinations discussed above, Table 13 also helps to identify countries that are just one step away from a suggested need for action, represented by countries G, H and I in the table. If Country H were to experience an increase either in the absolute prevalence or in the relative inequality by poverty, it could quickly fall into the fourth quartile on one of the two dimensions, thus being identified as a country with suggested priority for national action. A similar situation occurs for countries G and I, as they are in the fourth quartile for one dimension already, and would become a country with a suggested priority for national action if they fell into the third quartile for the other dimension. Thus, countries G, H and I may be considered “at risk”, since a deterioration of the situation could quickly result in their being categorized as having priorities for national action.

Using this approach, the figures and tables displayed in this assessment report were analysed to assess possible priorities for national action, applying the colour scheme defined in Table 13 to indicate countries with suggested priorities for national action (countries D, E and F), countries close to being classified as having suggested priorities for national action (countries G, H and I) and countries with a comparatively low relevance of inequalities.


Potential priorities for national action assessment findingsPossible priorities for inequality action are identified for the countries listed in Table 14, based on the merged analysis of relative and absolute inequality dimensions. However, it must be emphasized that a large number of countries cannot be assessed because no data are available, and therefore the identification of suggested priorities for national action is restricted to the respective and varying number of reporting countries for each given indicator. A more detailed table showing the assessment by country is available in Annex 3.

Interpretation of resultsIn countries with suggested priorities for action, national follow-up is needed to:• confirm and validate the findings based on national data;• identify the social and environmental context and causal mechanisms of the findings;• evaluate and interpret the respective inequalities.

This is especially important as the assessments presented in this report are based on quantitative data only, and might thus identify inequalities which, in a given national context, may not present a problem for country-specific reasons that cannot be considered by this international assessment report. Therefore, in these countries, action should first be focused on gaining a better understanding of the findings and the causal mechanisms leading to the observed disparities. Practical action should follow when the inequalities are confirmed as resulting from unfair societal processes.

In most cases, action would aim at the reduction of relative inequalities by reducing the exposure prevalence of the most affected population groups, which would also have an impact on the absolute exposure prevalence. In addition, countries might also consider general environmental protection measures to reduce the absolute population exposure, but would then need to pay attention to the distribution effects of the applied measures to ensure that more exposed population groups benefit as much from these efforts as the less exposed.

It must also be noted that this classification is based on a relative international comparison; thus individual countries may be performing more poorly than their neighbours, but may not have been identified as countries with potential priorities for action because other countries – many of which might have a lower developmental level – are reporting higher prevalence and inequality levels. Thus, if a similar classification approach were undertaken across only the EU15 countries, for example – a group more comparable in terms of economic situation than the group including all the Member States of the WHO European Region – some of these countries would have been categorized as countries with potential priorities for action, while they might not be categorized as such when data from a wider range of countries are taken into consideration. This is especially valid for “at risk” countries which, if categorized into a higher quartile for only one of the two dimensions, would immediately be classified as countries with suggested priorities for national action (see countries G, H and I in Table 13).

Furthermore, in some of the countries with small populations, the samples providing data for this assessment report were relatively small. This could have resulted in substantial random variability of the results, which must be borne in mind in their interpretation.

Finally, Table 14 reveals two important conclusions. First, for each environmental health inequality indicator, a large number of countries do not report the required data. For seven Member States, data were only available for 5 or fewer of the 30 assessed inequality dimensions covered within the 14 environmental health inequality indicators, and for five countries no data were identified for any of the inequality dimensions covered. In this context, environment and health policy-makers as well as experts from social support and protection sectors should bear in mind that the lack of data on the existing exposure differentials may represent an inequality in itself, hiding potential disparities between population groups behind national averages. Second, in a considerable number of countries the inequality trends are reversed and the environmental burden is higher in, for example, groups of higher


social position or income. This indicates that general trends say little about the situation in individual countries, and that for some environmental health risks, action may have to be targeted not only at the socially disadvantaged.

Table 14. Suggested priorities for national inequality action

Chapter Indicator relative inequality dimension Countries with suggested priorities for action

Country coverage

Housing Lack of flush toilet in dwelling

Above versus below relative poverty level

Bulgaria, Hungary, Latvia, Poland, Slovakia

30

Single-parent households versus all households

Austria [a], Bulgaria [a], Estonia [a], Greece [a]

30

Lack of bath or shower in dwelling

Lowest versus highest income quintile Belgium, Bulgaria, Cyprus, Estonia, Hungary, Latvia, Poland, Portugal, Romania

30

Single-parent households versus all households with children

Greece [a], Lithuania, Poland, Portugal, Slovenia

30

Overcrowding Single-parent households versus all households

Austria, Czech Republic 30

Lowest versus highest income quintile --- 30Dampness in the home

Lowest versus highest income quintile Bulgaria, Estonia, Latvia, Lithuania, Poland, Romania

30

Single-parent households versus all households

Cyprus, Poland, Romania 30

Inability to keep home warm in winter


Greece 30

Household type Cyprus, Germany, Poland 30Inability to keep home cool in summer

Lowest versus highest income quintile Cyprus, Greece, Italy, Portugal 27

Injury Work-related injuries

Male versus female Germany, Luxembourg, Portugal, Switzerland

15

Age group France, Portugal, Spain 15Mortality rate of all transport injuries

Age group San Marino 10

Mortality rate of RTIs Age group Croatia, Cyprus 37Male versus female Croatia, Italy, Lithuania, Portugal,

Serbia, Slovenia, Uzbekistan37

Poisoning mortality rate (all causes)

Male versus female Belarus, Estonia, Finland, Greece, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Poland, Russian Federation, Slovakia, Ukraine

43

Poisoning mortality rate (excluding alcohol poisoning)

Male versus female Estonia, Ireland, Kyrgyzstan, Latvia, Lithuania, Malta

32

Alcohol poisoning mortality rate

Male versus female Estonia, Finland, Kyrgyzstan, Latvia, Lithuania, Poland, Slovakia

28

Fall mortality rate Male versus female Belarus, Estonia, Finland, Latvia, Lithuania, Romania, Russian Federation

45

Environment Complaints about noise exposure at home


Denmark, Germany, Luxembourg, Netherlands, Romania [a]

30

Complaints about lack of access to recreational or green areas

Female versus male Poland [a], Portugal, Turkey 31Lowest versus highest income quartile Belgium, Bulgaria [a], Greece [a],

Hungary [a], Lithuania [a]31

No difficulty paying bills versus difficulty paying bills most of the time

--- 31

Exposure to second-hand smoke at home

Female versus male Luxembourg, the former Yugoslav Republic of Macedonia, Turkey

30

Low versus high self-assessed social position

Greece, Poland 30

No difficulty paying bills versus difficulty paying bills most of the time

--- 28

Exposure to second-hand smoke at work

Male versus female Austria, Lithuania 30Low versus high self-assessed social position

Bulgaria, Turkey [a] 28

Manager versus manual worker Cyprus, Italy, Malta, Spain 30

Notes: [a] reversed inequality dimension.Details of the country assessment can be found in Annex 3.

CONCLUSION

Conclusion130

CONCluSIONEditorial group

“Tackling the determinants of health inequalities is about tackling the unequal distribution of health determinants.”

(Graham, 2004)

KEy mESSAGES

There are four main conclusions from this assessment report.

1. Environmental health inequalities exist in all subregions and in all countries, and they are most often suffered by disadvantaged population groups.

• Environmental health inequalities exist across the region and are expressed by both intercountry and intracountry differences. Intracountry differences by social or demographic determinants can often exceed the intercountry differences, indicating that specific population groups are more strongly exposed to certain environmental risks within the respective country.

• For most of the environmental health inequality indicators for which stratification by socioeconomic determinants such as income or poverty is possible, the findings show a rather consistent pattern with socioeconomically disadvantaged individuals or households being more frequently exposed to environmental risks. However, in some countries, as well as for some risk factors, obverse inequalities can be found, indicating higher exposure levels in more affluent population groups.

• Environmental health inequality data can most often be stratified by age and gender, and frequently by socioeconomic determinants (mostly income-related, but also in relation to education or employment, for example) and household type. Each of these determinants is associated with significant inequalities.

º Income and poverty-related inequalities are identified for noise exposure, exposure to tobacco smoke at home and at work, and for housing-related inequality indicators, where they are expressed most strongly. Compared to the other social and demographic determinants applied, income- and poverty-related determinants display some of the strongest inequalities at regional as well as subregional and national levels (for EU27, the lack of a toilet in the home is, for example, eight times higher in the lowest income quintile than in the highest income quintile).

º Differences in national income levels are also associated with injury-related fatalities, with low/middle income countries reporting higher mortality rates.

º The indicators show that gender-related inequality is most strongly associated with injury, where male fatality rates are often three times (and beyond) female fatality rates. Gender differences also appear in relation to tobacco smoke exposure, yet play no important role for housing- and location-related risk factors.

º Age-related inequalities are present for injuries (especially falls) but are less prominent for the other inequality indicators.

º Household type-related inequalities are especially identified for single-parent households, and increase when combined with income-related determinants. Large variations by household type are identified for housing-related risk factors, putting vulnerable households at higher risk.

º Data on inequalities by education, employment/occupation level, self-reported social position and difficulty paying bills are only available for some of the environment-related inequalities, but show a more diverse inequality pattern. For example, high education level is consistently associated with lower self-reported access to recreational and green areas, while low social


position is most often associated with increased exposure to second-hand smoke both at home and at work. On the other hand, employment/occupation level shows different inequality patterns in exposure to second-hand smoke, depending on sex and subregion.

2. The magnitude of inequalities and the distribution of inequalities between advantaged and disadvantaged population groups can be very diverse between countries and also depends on the socioeconomic or demographic variable used for stratification.

• There are strong differences in the magnitude of environmental health inequalities between countries, and these differences are visible between countries both on disparate and on similar developmental levels. They are formed by both the differences in absolute exposure levels and the differences in the expression of relative inequalities, and show that each country provides a unique context for the development of environmental health inequalities.

• For a range of inequality dimensions, some countries exhibit opposing inequality directions compared to the majority of countries. This variation of inequality direction was often found for inequalities by sex and age, but also occurred – depending on the environmental risk considered – in relation to inequalities by income and poverty, by household type, and other determinants such as social position, difficulty paying bills, and occupation.

3. To allow reliable identification of the most relevant target groups and to understand better the national inequality patterns and their causal mechanisms, more detailed environmental health inequality reporting and assessment must be carried out at the national level.

• Large intercountry variations and national exceptions from regional or subregional trends exist for many indicators. More detailed environmental health inequality assessment should be carried out at a national or subnational scale, as general patterns for the WHO European Region or its subregions are not necessarily reflected in all countries.

• The general population exposure to environmental risks does not necessarily indicate the level of relative inequalities. For some indicators, inequality expressed in relative terms is higher in countries with rather modest general exposure prevalence. This situation often applies to the more affluent countries due to the very low prevalence of environmental risks in the advantaged population subgroups (such as those with the highest SES).

• The use of national databases for country-specific environmental health inequality assessments will enable more detailed findings by providing additional data on exposure as well as enabling analysis of a more complete set of social and/or demographic determinants.

4. The evidence base for the assessment of environmental health inequalities is weak and needs to be strengthened.

• The overall availability of reliable and consistent national and international data capable of indicating differences by social or demographic determinants is insufficient. In particular, a large evidence gap exists between EU Member States, EU candidate countries and European Free Trade Association countries on the one hand, and the other countries in the WHO European Region on the other. This evidence gap becomes especially apparent when it is noted that of the 14 environmental health inequality indicators covered in this report, only 5 can draw data from the whole WHO European Region, while consistent data for all 14 indicators are available from the EU countries.

• Many environmental health inequality indicators covered in this report are restricted to stratification by only one or two sociodemographic determinants, and for a number of environmental health priorities (such as chemical exposure, air quality, distance to waste sites or location in dangerous residential areas) no inequality indicators could be developed because of a lack of data.

• Several of the indicators used are based on self-reported data and have been taken from national surveys with relatively small sample sizes.

Conclusion132

In summary, this report compiles and quantifies a diverse range of environmental health inequalities, proving that challenges of environmental justice are apparent in all countries, and documenting that each of the environmental inequalities covered is associated with a range of health effects, thus contributing to health inequalities. However, the data are rather limited, both in terms of quantity (including low country coverage, lack of data on risk factors and on sociodemographic determinants) and sometimes quality (such as small sample sizes and dependency on self-reported or vague exposure conditions). Therefore, this assessment report marks only a first milestone in the progress towards a reliable and comprehensive assessment of environmental health inequality in the WHO European Region.

Furthermore, it is not possible without further investigation to say whether the observed differences in environmental exposures between different sociodemographic groups are simply the product of chance, or whether they reflect the inequities discussed in Chapter 1: inequalities resulting from a combination of influences, which imply unfairness and injustice and are avoidable. A qualitative assessment of whether the differences described in this report are unavoidable inequalities or unfair reflections of inequity is a task best conducted at the national level, where the particular social and policy context is better understood. Therefore, the results of this report should encourage Member States and their governments to follow up in more detail on the findings and inequalities presented. This should permit the identification of areas where action is most needed and where it is possible to reduce the described inequalities.

FurTHEr PErSPECTIvES FOr ACTIONThe findings of the environmental health inequality assessment report indicate that, although to differing extents, environmental health inequalities exist in all Member States of the WHO European Region. Many countries need to tackle environmental health inequalities as a priority issue, although national priorities and disadvantaged groups vary. Six general recommendations for action that can be derived from this report are given below, but need to be tailored to the national situation.

Action 1: general improvement of environmental conditionsIrrespective of target groups and target areas, general improvement of housing conditions, safety regulations and education, and environmental conditions would reduce the general exposure level of all citizens, thereby also improving the situation of those most affected or most vulnerable. Disadvantaged population groups may even benefit to a greater extent from such interventions, and thus basic measures providing adequate environmental conditions to all would be likely to mitigate a range of environmental health inequalities.

As indicated in the Introduction (Fig. 1), the expression of inequalities as gradients across population subgroups may offer helpful information in deciding whether universal action for the benefit of all is a suitable strategy for tackling these inequalities. This report enables the assessment of environmental health issues affecting most or all population groups, which call for universal actions to create healthy environments for all. Such general interventions could be suggested, for example, to address:• water supply in all rural areas of Euro 3 countries (see Fig. 3, Chapter 2);• overcrowding in EU15 and NMS12 countries (see Fig. 11, Chapter 2);• dampness in dwellings in EU15 and NMS12 countries (see Fig. 14, Chapter 2);• inability to keep the home cool in summer in EU15 and NMS12 countries (see Fig. 18, Chapter 2);• noise exposure in EU15 and NMS12 countries (see Fig. 34, Chapter 4).

However, as environmental conditions and inequalities may be diverse, more detailed national assessment on general environmental priorities is recommended before taking action (see Action 3 below).


Action 2: mitigation and reduction of risk exposure in the most affected population groupsTargeted and short-term action is necessary to reduce exposure burdens in the most exposed as well as the most vulnerable population groups. If specific inequality assessments to identify the most affected groups are not available, area-based approaches could be applied. These could focus on environmental exposure hotspots or neighbourhoods/residential locations known to be most deprived and/or exposed to housing and environmental risks, including transport-related exposures such as noise, air pollution and safety threats.

As with identification of universal actions, some findings indicate that specific target groups are most at risk. Acknowledging that especially steep or skewed gradients suggest the application of risk group-specific interventions, this report proposes that targeted action could be taken, for example, to address:• lack of flush toilet in low-income households in NMS12 countries (see Fig. 7, Chapter 2);• lack of bath or shower in low-income households in NMS12 countries (see Fig. 9, Chapter 2);• RTI mortality in young male adults across all regions (see Fig. 26, Chapter 3);• lack of access to recreational/green areas in population with high education level in Euro 4 countries

(see Table 10, Chapter 4).

Again, within individual countries, gradients and inequalities may be diverse or reversed, so more detailed national assessment is recommended before taking targeted action (see Action 3 below).

Action 3: national environmental health inequality assessmentsThis report shows that the profile of environmental health inequalities is very different across Member States. To allow targeted action, further country-specific work is necessary to collect more and better data on national priorities and the population groups most at risk. This could also be achieved by integration of inequality-sensitive items in national health and/or environment surveys, or the development of multiple indices which bring the data together to provide a more holistic measure. The methodology and environmental health inequality indicators developed during this project could provide a backbone for country-specific adaptations of the WHO European environmental health inequality assessment. Furthermore, an effort towards better data collection standardization would allow more useful and reliable international comparisons.

Action 4: sharing experiences – case studies on successful interventionsA number of countries have already provided evidence on their experiences of identifying and assessing environmental health inequalities, as well as the actions and interventions used to tackle those inequalities (see Annex 2). A review of case studies based on interventions in the health sector as well as in non-health sectors (including environment, transport, urban planning, work, education, social and housing) could identify successful approaches and elements of good practice for the reduction of environmental health inequalities related to, for example:• enforcement of legislation (such as smoking bans, speed limits and building codes);• integration of an inequality perspective into urban planning (such as safe design of buildings and

activity-friendly neighbourhoods, and safe and ecological transport modes);• integration of an inequality perspective into policy measures aimed at the reduction and equal

distribution of environmental hazards;• targeted environmental, social and infrastructural measures to protect the most vulnerable groups.

Conclusion134

Action 5: review and modification of national intersectoral policies in relation to environmental health inequalitiesReflecting the fact that environmental and health inequalities are often caused by policies and decisions made outside the environment and health sectors, increased communication and exchange between sectors are necessary to ensure shared information and to prevent the negative consequences of equity-sensitive policies and measures. Moderated round-table discussions could bring together high-level representatives of different sectors to discuss and agree on opportunities to reduce unequal distributive effects of policy-making. In this context, briefing papers could be developed that inform non-health sectors about the health and equity consequences of their actions.

Action 6: monitoring of environmental health inequalitiesWhile single assessments and status reports on environmental health inequalities will be useful to provide evidence and raise awareness, continuous monitoring of environmental health inequalities is necessary to trace the trends, re-assess priorities and evaluate the success of policy or technical interventions. Again, national action in monitoring the development of environmental health inequalities will be most effective. Nevertheless, WHO will make use of its Environment and Health Information System (ENHIS) as an existing structure to monitor environment and health trends, and extend selected ENHIS indicators to cover stratification by relevant social or demographic determinants as respective data become available.

rEFErENCE

Graham H (2004). Tackling inequalities in health in England: remedying health disadvantages, narrowing health gaps or reducing health gradients? Journal of Social Policy, 33(1):115–31.

1

NATIONAL ENVIRONMENTAL

HEALTH INEQUALITY FACT SHEETS

ANNEX

Annex 1. National environmental health inequality fact sheets136

ANNEx 1. NATIONAl ENvIrONmENTAl HEAlTH INEquAlITy FACT SHEETS

CONTENTS

Dwellings supplied with piped water by proportion of Roma population (Hungary) ��137

Lack of flush toilet by wealth status, urban/rural residence and region (Georgia) ��138

Lack of bath or shower in dwelling by urban/rural population and region (Kyrgyzstan) ��139

Overcrowding by tenure group and household income (Great Britain) ��140

Dampness in dwelling by age, income and household type (Norway) ��141

Lack of home heating by education and household composition (Serbia and Montenegro) ��142

Work-related injuries by sex and economic sector (Croatia) ��143

Transport-related mortality by sex and age (Malta) ��144

Mortality from accidental poisoning by age and urban/rural residence (Poland) ��145

Mortality from falls by age and sex (Romania) ��146

Exposure to traffic noise within residential areas by household income level (Netherlands) ��147

Lack of access to recreational/green space by region (Spain) ��148

Potential smoke exposure at home by SES (Germany) ��149

Exposure to second-hand smoke at work by sex, age, income, and employment (Italy) ��150


dw

ellin

gs s

uppl

ied

wit

h pi

ped

wat

er b

y pr

opor

tion

of r

oma

popu

lati

on

KE

Y M

ESSA

GE:

The

per

cent

age o

f dwe

lling

s sup

plied

with

pip

ed w

ater

diff

ers b

etwe

en

mun

icipa

lities

in H

unga

ry w

ith d

iffer

ent p

ropo

rtion

s of R

oma p

opul

atio

n.

Dat

a so

urce

: The

dat

a ar

e ta

ken

from

the

200

1 ce

nsus

und

erta

ken

by t

he H

unga

rian

Cen

tral

Stat

istica

l Offi

ce a

nd s

tore

d at

the

Reg

iona

l Inf

orm

atio

nal S

yste

m o

f th

e M

inist

ry o

f N

atio

nal

Dev

elopm

ent.

Dat

a de

scrip

tion:

The

par

amet

ers

are

defin

ed a

s “d

welli

ngs

supp

lied

with

pip

ed w

ater

” and

as

“pro

porti

on o

f Rom

a pop

ulat

ion

in m

unici

palit

y”.

Dat

a co

vera

ge: T

he a

vaila

ble

data

hav

e na

tiona

l cov

erag

e, str

atifi

ed b

y m

unici

palit

y lev

el (L

ocal

Adm

inist

rativ

e Uni

t2, f

orm

erly

know

n as

NU

TS5

).

Soci

al d

imen

sion:

Thi

s in

equa

lity

fact

she

et d

escr

ibes

“dw

ellin

gs s

uppl

ied w

ith p

iped

wat

er” i

n as

socia

tion

with

the p

ropo

rtion

of R

oma p

opul

atio

n in

the r

espe

ctive

mun

icipa

lity (

divid

ed in

to fi

ve

grou

ps b

y nat

ural

brea

ks in

the d

ata)

.

Con

text

: The

Rom

a po

pulat

ion

is th

e on

ly of

ficial

eth

nic g

roup

in H

unga

ry. T

he ce

nsus

def

initi

on

of a

Rom

a inh

abita

nt is

bas

ed o

n a p

erso

nal d

eclar

atio

n of

bein

g a R

oma (

aggr

egat

ed d

escr

iptio

n of

th

e Gip

sy, R

oman

y and

Bea

pop

ulat

ion)

.

The

Rom

a pop

ulat

ion

is re

lative

ly sm

all, c

ompr

ising

onl

y 2.0

2% o

f the

tota

l pop

ulat

ion

acco

rdin

g to

th

e 200

1 ce

nsus

in H

unga

ry. H

owev

er, th

e pro

porti

on is

asse

ssed

as 7

% in

the E

U st

rate

gy o

n Ro

ma

inclu

sion.

Rom

a in

habi

tant

s live

main

ly in

the

less u

rban

ized

, rur

al ar

eas w

here

the

popu

latio

n ha

s a m

ixed

ethn

ic–re

ligio

us c

ompo

sitio

n. H

igh

unem

ploy

men

t rat

es a

nd a

low

educ

atio

nal l

evel

char

acte

rize

this

ethn

ic gr

oup.

The

nat

iona

l av

erag

e of

dwe

lling

s su

pplie

d wi

th p

iped

wat

er i

s 90

.7%

but

can

be

belo

w 50

%

depe

ndin

g on

the r

espe

ctive

mun

icipa

lity.

Polic

y im

plic

atio

ns: E

arlie

r sur

veys

hav

e sho

wn th

at ap

prox

imat

ely 1

.6%

of t

he to

tal p

opul

atio

n of

Hun

gary

live

s in

segr

egat

ed h

abita

ts (c

olon

ies) w

ith p

oor e

nviro

nmen

tal c

ondi

tions

su

ch as

a lac

k of

sewe

rage

and

gas s

uppl

y, th

e pre

senc

e of g

arba

ge d

epos

its an

d wa

terlo

gged

soil.

94%

of t

hese

colo

nies

are i

nhab

ited

pred

omin

antly

by R

oma.

The

mos

t dep

rived

area

s ar

e tho

se p

arts

of th

e cou

ntry

with

the h

ighe

st nu

mbe

rs of

colo

nies

, and

cons

eque

ntly

a hig

her p

ropo

rtion

of R

oma p

opul

atio

n.

Sign

ifica

nt im

prov

emen

ts of

livin

g co

nditi

ons

can

be a

chiev

ed b

y im

plem

entin

g co

mpl

ex p

rogr

amm

es to

cha

nge

envir

onm

enta

l fac

tors

at lo

cal a

nd n

atio

nal l

evels

. The

se p

lanne

d pr

ogra

mm

es w

ill b

e in

acco

rdan

ce w

ith th

e main

aim

s of t

he n

ewly

adop

ted

EU st

rate

gy o

n Ro

ma i

nclu

sion

to cl

ose t

he g

ap in

acce

ss to

hou

sing

and

publ

ic ut

ilitie

s suc

h as

wat

er an

d ele

ctric

ity (E

U F

ram

ewor

k fo

r Nat

iona

l Rom

a Int

egra

tion

Stra

tegi

es u

p to

202

0, 5.

4.20

11 C

OM

(201

1) 1

73 fi

nal).Dw

ellin

gs s

uppl

ied

with

pip

ed w

ater

by

prop

ortio

n of

Rom

a po

pula

tion

0102030405060708090100

0–3.

13.

2–9.

910

–21.

121

.2–4

2.7

42.8

–93.

4

Perc

enta

ge o

f Rom

a po

pula

tion

in m

unic

ipal

ity

Percentage of dwellings supplied with piped water

Nat

iona

l ave

rage

leve

l (90

.7%

)

Ana

lysis

and

inte

rpre

tatio

n: A

stro

ng, s

igni

fican

tly in

verse

ass

ociat

ion

is fo

und

betw

een

betw

een

the p

erce

ntag

e of d

welli

ngs s

uppl

ied w

ith p

iped

wat

er an

d th

e pro

porti

on o

f Rom

a pop

ulat

ion

at th

e m

unici

pal l

evel

in H

unga

ry, b

ased

on

the 2

001

cens

us.

The

ban

d sh

owin

g th

e low

est p

erce

ntag

e of d

welli

ngs s

uppl

ied w

ith p

iped

wat

er is

char

acte

rized

by

the h

ighe

st pr

opor

tion

of R

oma p

opul

atio

n. In

this

band

less

than

one-

half

(45.

29%

) of d

welli

ngs a

re

supp

lied

with

pip

ed w

ater,

a fig

ure 5

0% lo

wer t

han

the n

atio

nal a

vera

ge (9

0.7%

).

Hun

gary


KE

Y M

ESSA

GE:

Hou

seho

lds w

ith lo

w we

alth

statu

s lev

els ar

e muc

h m

ore l

ikely

to li

ve

in d

welli

ngs

with

out f

lush

toile

ts in

Geo

rgia.

Rur

al ar

eas

are

espe

cially

affe

cted

by

this

ineq

ualit

y.

Dat

a sou

rce:

Dat

a on

the a

vaila

bilit

y of f

lush

toile

ts ar

e tak

en fr

om th

e Rep

rodu

ctive

Hea

lth S

urve

y ca

rried

out

in G

eorg

ia in

201

0 jo

intly

by th

e Nat

iona

l Cen

ter f

or D

iseas

e Con

trol a

nd P

ublic

Hea

lth,

the

Cen

ters

for

Dise

ase

Con

trol a

nd P

reve

ntio

n, t

he U

nite

d N

atio

ns P

opul

atio

n Fu

nd a

nd t

he

Uni

ted

Stat

es A

genc

y for

Inte

rnat

iona

l Dev

elopm

ent.

Dat

a de

scrip

tion:

The

hou

seho

ld c

hapt

er o

f th

e su

rvey

def

ines

var

ious

par

amet

ers,

inclu

ding

cla

ssifi

catio

ns o

f “flu

sh to

ilet”

and

“pit

latrin

e” as

pre

-cod

ed o

ptio

ns, a

mon

g ot

her a

ltern

ative

s, an

d re

fers

to h

ouse

hold

s.

Dat

a cov

erag

e: T

he av

ailab

le da

ta h

ave n

atio

nal c

over

age a

nd ar

e ava

ilabl

e for

201

0. Fu

ture

surv

eys

are p

lanne

d.

Soci

al d

imen

sion:

Thi

s ine

quali

ty fa

ct sh

eet s

tratif

ies “a

vaila

bilit

y of f

lush

toile

t” by

wea

lth q

uint

ile,

resid

ence

and

selec

ted

regi

ons.

Wea

lth st

atus

is b

ased

on

hous

ehol

d as

sets,

inclu

ding

dur

able

good

s (su

ch a

s re

frige

rato

r, te

levisi

on, c

ar a

nd c

ompu

ter)

and

dwell

ing

char

acte

ristic

s (su

ch a

s ty

pe o

f so

urce

for

drin

king

-wat

er, t

oilet

fac

ilitie

s, fu

el us

ed f

or c

ooki

ng a

nd h

eatin

g, m

ain r

oof

mat

erial

an

d ov

ercr

owdi

ng).

The

sam

ple

of h

ouse

hold

s was

divi

ded

into

qui

ntile

s of e

qual

size

base

d on

a

weig

hted

freq

uenc

y dist

ribut

ion

of h

ouse

hold

s by t

he re

sulti

ng as

set s

core

indi

catin

g we

alth

statu

s.

Con

text

: In

gene

ral,

48%

of a

ll G

eorg

ian h

ouse

hold

s ha

ve a

flus

h to

ilet w

ith a

dequ

ate

hygi

enic

stand

ards

, whi

le 50

% h

ave

a pi

t lat

rine

that

pro

vides

a m

uch

less

appr

opria

te h

ygien

ic co

ntex

t. H

owev

er, th

e ava

ilabi

lity o

f a fl

ush

toile

t is h

ighl

y ass

ociat

ed w

ith u

rban

setti

ngs a

nd in

crea

sed

socia

l sta

tus a

nd w

ealth

: alm

ost t

hree

out

of f

our u

rban

hou

seho

lds (

74%

) are

clas

sified

in th

e two

hig

hest

wealt

h qu

intil

es, w

hile

only

3% o

f rur

al ho

useh

olds

are

as w

ealth

y. In

stead

, mor

e th

an 7

0% o

f rur

al ho

useh

olds

belo

ng to

the l

ow an

d lo

west

wealt

h qu

intil

e.

In c

onse

quen

ce, t

he a

vaila

bilit

y of

flu

sh t

oilet

s ve

rsus

pit

latrin

es f

ollo

ws a

stro

ng s

ocial

and

ge

ogra

phica

l pat

tern

, put

ting

rura

l and

less

dev

elope

d re

gion

s with

less

eco

nom

ic po

wer a

nd m

ore

tradi

tiona

l hou

sing

stock

at a

disa

dvan

tage

.

Polic

y im

plic

atio

ns: T

here

is a

ver

y str

ong

diffe

renc

e in

the

avail

abili

ty o

f flu

sh to

ilets

betw

een

hous

ehol

ds in

Geo

rgia

by w

ealth

qui

ntile

, as

well

as b

y ur

ban/

rura

l res

iden

ce a

nd

geog

raph

ic ar

ea. L

owes

t wea

lth q

uint

ile h

ouse

hold

s exc

lusiv

ely li

ve in

dwe

lling

s with

out f

lush

toile

ts, w

hich

are a

pre

dom

inan

tly ru

ral f

eatu

re, w

hile

urba

n ho

useh

olds

stro

ngly

bene

fit

from

econ

omic

wealt

h an

d th

us h

ave a

very

hig

h lev

el of

acce

ss to

flus

h to

ilets.

Pol

itica

l mea

sure

s to

redu

ce th

e une

qual

distr

ibut

ion

of h

ygien

ic sa

nita

tion

equi

pmen

t the

refo

re n

eed

to

be ta

ckled

pre

dom

inan

tly th

roug

h ec

onom

ic m

easu

res a

nd an

impr

ovem

ent o

f livi

ng co

nditi

ons i

n ru

ral a

reas

.Ava

ilabi

lity

of fl

ush

toile

t by

wea

lth s

tatu

s

020

4060

8010

0

Low

est

Low

Mid

dle

Hig

h

Hig

hest

Wealth quintile

Perc

enta

ge o

f pop

ulat

ion

Ava

ilabi

lity

of to

ilet t

ype

by u

rban

/ru

ral r

esid

ence

and

sel

ecte

d re

gion

020

4060

8010

0

Urb

an

Rur

al

Tbilis

i

Adja

ra

Kvem

o Ka

rtli

Kakh

eti

Rac

ha-S

vane

ti

Perc

enta

ge o

f pop

ulat

ion

Flus

h to

ilet

Pit l

atrin

eO

ther

toile

t fac

ilitie

s

Ana

lysis

and

inte

rpre

tatio

n: T

here

is a

ver

y str

ong

diffe

renc

e in

the

ava

ilabi

lity

of fl

ush

toile

ts be

twee

n ho

useh

olds

by w

ealth

qui

ntile

, ran

ging

from

0–1

00%

. Sim

ilarly

, stro

ng d

ispar

ities

rega

rdin

g th

e ava

ilabi

lity o

f flu

sh to

ilets

exist

by u

rban

/rura

l res

iden

ce an

d ge

ogra

phic

area

. Ove

rall,

the l

owes

t we

alth

quin

tile h

ouse

hold

s live

exclu

sively

in d

welli

ngs w

ithou

t a fl

ush

toile

t.

The

situ

atio

n is

mos

t sev

ere i

n ru

ral a

reas

(aro

und

10%

of h

ouse

hold

s hav

ing

acce

ss to

flus

h to

ilets)

an

d pa

rticu

larly

in m

ount

ainou

s are

as (s

uch

as th

e Rac

ha an

d Sv

anet

i reg

ions

, at 8

%).

The

une

qual

distr

ibut

ion

indi

cate

s th

e de

gree

to w

hich

wea

lth is

dist

ribut

ed in

geo

grap

hic

area

s, de

mon

strat

ing

a di

rect

effe

ct o

f pov

erty

and

eco

nom

ic di

sadv

anta

ge o

n ba

sic sa

nita

ry e

quip

men

t. Lo

okin

g at

the r

egio

nal v

ariat

ion,

Tbi

lisi m

etro

polit

an ar

ea h

as th

e lar

gest

prop

ortio

n of

hou

seho

lds

in th

e hig

hest

two

wealt

h qu

intil

es (9

1%) a

nd th

us th

e hig

hest

level

of fl

ush

toile

t ava

ilabi

lity (

96%

), wh

ile re

gion

s with

mos

t hou

seho

lds i

n th

e low

est t

wo w

ealth

qui

ntile

s rep

ort t

hat 6

0% an

d m

ore o

f th

eir d

welli

ngs a

re o

nly e

quip

ped

with

pit

latrin

es.

lack

of f

lush

toi

let

by w

ealt

h st

atus

, urb

an/r

ural

res

iden

ce a

nd r

egio

nG

eorg

ia


KE

Y M

ESSA

GE:

In K

yrgy

zsta

n lar

ger p

ropo

rtion

s of r

ural

than

urb

an p

opul

atio

ns la

ck a

bath

or s

howe

r in

their

dwe

lling

. The

nat

iona

l rat

io is

1.7

:1, w

ith re

gion

al va

riatio

ns.

Dat

a sou

rce:

Dat

a on

the a

vaila

bilit

y of a

bat

h or

show

er in

the d

welli

ng ar

e tak

en fr

om th

e nat

iona

l ce

nsus

, con

duct

ed ev

ery 1

0 ye

ars b

y the

Nat

iona

l Sta

tistic

al C

omm

ittee

of K

yrgy

zsta

n. T

he d

ata u

sed

are f

rom

the 1

999

cens

us. S

umm

ary d

ata o

f the

200

9 ce

nsus

can

be d

ownl

oade

d fro

m th

e web

site

at:

http

://ww

w.sta

t.kg,

but h

ousin

g am

eniti

es d

ata a

re st

ill u

nder

analy

sis.

Dat

a de

scrip

tion:

The

par

amet

er i

s de

fined

as

“bat

h (sh

ower

) in

dwe

lling

”, co

rresp

ondi

ng t

o dw

ellin

gs in

whi

ch a

bath

or a

show

er is

insta

lled

eithe

r in

a sep

arat

e bat

hroo

m o

r in

any o

ther

room

fit

ted

for t

his p

urpo

se, ir

resp

ectiv

e of t

he ty

pe o

f sup

ply o

f hot

wat

er.

Dat

a cov

erag

e: T

he su

rvey

has

nat

iona

l cov

erag

e, str

atifi

ed b

y sev

en re

gion

s and

the c

ity o

f Bish

kek.

Dat

a are

avail

able

for 1

999.

Soci

al d

imen

sion:

Thi

s ine

quali

ty fa

ct sh

eet s

tratif

ies th

e per

cent

age o

f urb

an an

d ru

ral p

opul

atio

ns

with

no

bath

or s

howe

r in

the d

efin

ed re

gion

al en

tities

.

Con

text

: The

199

9 po

pulat

ion

cens

us o

f Kyr

gyzs

tan

reco

rded

1 1

09 0

00 h

ouse

hold

s, wi

th 4

787

000

re

siden

ts; 1

35 h

ouse

hold

s (25

3 pe

ople)

wer

e hom

eless

. The

cens

us co

ntain

s inf

orm

atio

n on

the l

evel

of h

ousin

g am

eniti

es in

the c

ount

ry. W

here

as el

ectri

city i

s ava

ilabl

e to

prac

ticall

y all

the p

opul

atio

n,

man

y hou

ses –

espe

cially

thos

e not

bui

lt by

the s

tate

– la

ck ru

nnin

g wa

ter, c

entra

l hea

ting,

hot w

ater

su

pply,

bat

h or

show

er, g

as an

d se

wera

ge co

nnec

tions

.

In 1

989

73%

of

hous

ing

was

owne

d by

citi

zens

, whe

reas

acc

ordi

ng t

o th

e 19

99 c

ensu

s ho

me-

owne

rship

has

rise

n to

94.

6%. T

his

rise

occu

rred

due

to th

e pr

ivatiz

atio

n of

sta

te-o

wned

hou

ses,

and

has i

ncre

ased

the l

evel

of p

rivat

e res

pons

ibili

ty fo

r hou

sing

amen

ities

such

as h

ygien

e fac

ilitie

s.

Polic

y im

plic

atio

ns: T

here

is a

ver

y str

ong

diffe

renc

e in

ava

ilabi

lity

of a

bat

h or

show

er b

etwe

en u

rban

(42%

) and

rura

l (2.

1%) p

opul

atio

ns in

Kyr

gyzs

tan.

In th

e cit

y of

Bish

kek

the

popu

latio

n liv

es in

hou

ses w

ith a

high

er le

vel o

f am

eniti

es th

an th

e pop

ulat

ion

of o

ther

area

s, in

dica

ted

by av

ailab

ility

of a

bat

h or

show

er fo

r 57.

6% o

f the

urb

an an

d 6.

4% fo

r the

rura

l po

pulat

ion.

At p

rese

nt, f

ollo

wing

priv

atiz

atio

n, th

e m

ajorit

y of

hou

sing

stock

is in

priv

ate

owne

rship

. Acc

ess t

o ne

w ho

uses

with

am

eniti

es d

epen

ds o

n th

e so

cioec

onom

ic lev

el of

the

hous

ehol

d.

Dat

a fro

m th

e 200

9 ce

nsus

will

help

to id

entif

y reg

ions

requ

iring

urg

ent a

ttent

ion

rega

rdin

g lac

k of

a ba

th o

r sho

wer. G

uida

nce o

n th

e im

prov

emen

t of h

ousin

g co

nditi

ons i

s inc

lude

d in

na

tiona

l hou

sing

regu

latio

ns. H

owev

er, sa

nita

ry an

d ep

idem

iolo

gica

l req

uire

men

ts fo

r res

iden

tial h

ouse

s and

indo

or sp

aces

are c

urre

ntly

only

reco

mm

enda

tions

; tec

hnica

l reg

ulat

ions

are

yet t

o be

dev

elope

d. D

evelo

pmen

t and

revis

ion

of g

ener

al bu

ildin

g an

d de

velo

pmen

t plan

s of c

ities

and

resid

entia

l are

as sh

ould

take

into

cons

ider

atio

n ne

w ty

pes o

f bui

ldin

g, in

cludi

ng

low-

rise

and

indi

vidua

lly d

esig

ned

dwell

ings

. Iss

ues o

f pro

vidin

g ex

istin

g dw

ellin

gs w

ith sa

nita

ry a

men

ities

(inc

ludi

ng a

bat

h or

show

er) a

nd th

e ne

cess

ary

infra

struc

ture

nee

d to

be

reso

lved.

The

re is

a ne

ed to

impr

ove a

nd d

evelo

p te

chni

cal r

egul

atio

ns o

n ho

usin

g in

ord

er to

ensu

re sa

fe an

d co

mfo

rtabl

e con

ditio

ns fo

r the

pop

ulat

ion

of K

yrgy

zsta

n.

Lack

of b

ath

or s

how

er b

y ur

ban/

rura

l pop

ulat

ion

and

regi

on

0102030405060708090100

Kyrgyzstan

Batkenregion

Djalal-Abadregion

Issyk-Kulregion

Narynregion

Oshregion

Talas region

Chuiregion

Bishkekcity

Percentage of population

Urb

anR

ural

Ana

lysis

and

inte

rpre

tatio

n: O

nly

15.9

% o

f the

pop

ulat

ion

of K

yrgy

zsta

n ha

ve a

bat

h or

show

er

in th

eir d

welli

ng. H

owev

er, th

ere

is str

ong

urba

n–ru

ral i

nequ

ality,

as 5

8% o

f the

urb

an p

opul

atio

n ve

rsus 9

7.9%

of t

he ru

ral p

opul

atio

n do

not

hav

e a b

ath

or sh

ower.

Sim

ilarly

, stro

ng d

iffer

ence

s exis

t by

regi

on. A

mon

g ur

ban

area

s the

lowe

st pe

rcen

tage

of p

opul

atio

n ha

ving

no b

ath

or sh

ower

is in

th

e Chu

i reg

ion

(58.

2%) a

nd th

e hig

hest

is in

the T

alas r

egio

n (9

6.8%

). A

mon

g rur

al ar

eas t

he lo

west

perc

enta

ge o

f the

pop

ulat

ion

havin

g no

bat

h or

show

er is

also

in th

e Chu

i reg

ion

(91.

1%),

while

the

high

est i

s in

the O

sh an

d th

e Nar

yn re

gion

s (99

.8%

). The

hig

hest

rate

s of a

cces

s to

a bat

h or

show

er

are i

n th

e cap

ital c

ity, B

ishke

k.

In K

yrgy

zsta

n lac

k of

a b

ath

or sh

ower

in th

e dw

ellin

g is

muc

h m

ore

stron

gly

evid

ent i

n th

e ru

ral

than

the u

rban

pop

ulat

ion

(at a

ratio

of 1

.7:1

). T

he d

iffer

ence

bet

ween

regi

onal

prop

ortio

ns o

f rur

al an

d ur

ban

popu

latio

ns la

ckin

g a b

ath

or sh

ower

is lo

west

in th

e Tala

s and

Nar

yn re

gion

s.

lack

of b

ath

or s

how

er in

dw

ellin

g by

urb

an/r

ural

pop

ulat

ion

and

regi

onKy

rgyz

stan


KE

Y M

ESSA

GE:

The

hig

hest

prop

ortio

ns o

f ove

rcro

wdin

g fo

r all

inco

me

quin

tiles

are

fo

und

in th

e so

cial r

ente

d se

ctor

. Ove

rcro

wdin

g af

fect

s all

tenu

re g

roup

s but

in e

ach

case

is

mor

e pre

valen

t in

the t

hree

lowe

st in

com

e qui

ntile

s. Po

licies

that

targ

et o

verc

rowd

ing

in

the s

ocial

rent

ed se

ctor

hav

e the

pot

entia

l to

bene

fit m

ore f

amili

es.

Dat

a so

urce

s: D

ata

on o

verc

rowd

ing

are

avail

able

from

var

ious

sou

rces

inc

ludi

ng t

he G

ener

al Li

festy

le Su

rvey

(GLS

), fo

rmer

ly kn

own

as th

e Gen

eral

Hou

seho

ld S

urve

y (ht

tp://

www.

ons.g

ov.u

k/on

s/sea

rch/

inde

x.htm

l?pag

eSiz

e=50

&ne

wque

ry=g

ener

al+lif

esty

le+su

rvey

).

Dat

a de

scrip

tion:

The

GLS

par

amet

er c

hose

n to

ass

ess

over

crow

ding

is b

ased

on

the

‘bed

room

sta

ndar

d’: t

he n

umbe

r of b

edro

oms r

equi

red

in a

hous

e is c

alcul

ated

from

the n

umbe

r of p

eopl

e livi

ng

ther

e an

d th

eir re

latio

nshi

ps to

eac

h ot

her.

One

bed

room

is a

lloca

ted

for e

ach

mar

ried/

coha

bitin

g co

uple,

for a

ny o

ther

per

son

aged

ove

r 21,

for e

ach

sam

e-se

x pa

ir ag

ed 1

0–20

and

for e

ach

pair

of

child

ren

unde

r 10.

Whe

re th

ere a

re o

ne o

r mor

e bed

room

s too

few

for t

he o

ccup

ants

of th

e hou

se it

is

said

to b

e ove

rcro

wded

.

Dat

a co

vera

ge: T

he G

LS d

ata

used

are

for

the

year

s 20

04 to

200

6. T

he o

vera

ll sa

mpl

e wa

s 31

23

3 ho

useh

olds

(oc

cupi

ed b

y 74

314

peo

ple)

. Sam

plin

g fo

r th

e G

LS is

bas

ed o

n a

prob

abili

ty-

strat

ified

two-

stage

sam

ple

desig

n an

d is

repr

esen

tativ

e of

Gre

at B

ritain

. The

dat

a ar

e we

ight

ed to

th

e pop

ulat

ion

and

to ta

ke ac

coun

t of n

on-r

espo

nse a

nd at

tritio

n. N

orth

ern

Irelan

d is

not i

nclu

ded

in th

e cov

erag

e of t

he su

rvey

.

Soci

al d

imen

sion:

Thi

s ine

quali

ty fa

ct sh

eet s

tratif

ies o

verc

rowd

ing

by te

nure

and

inco

me q

uint

ile.

The

tenu

re ca

tego

ries a

re o

wner

-occ

upat

ion,

rent

al fro

m a

socia

l lan

dlor

d (su

ch as

a lo

cal a

utho

rity)

an

d re

ntal

from

a pr

ivate

land

lord

.

Con

text

: Ove

rcro

wdin

g ha

s hist

orica

lly b

een

linke

d to

incr

ease

d tra

nsm

issio

n of

infe

ctio

us d

iseas

e. In

add

ition

, ove

rcro

wded

con

ditio

ns a

re n

ow s

een

as d

etrim

enta

l to

educ

atio

nal a

ttain

men

t – fo

r ex

ampl

e, by

den

ying

you

ng p

eopl

e th

e op

portu

nity

for

uni

nter

rupt

ed s

tudy

. For

adu

lts, a

nxiet

y an

d de

pres

sion

can

resu

lt fro

m c

rowd

ed c

ondi

tions

and

lack

of

oppo

rtuni

ty f

or p

rivac

y. W

hile

over

crow

ding

leve

ls ha

ve fa

llen

in th

e las

t few

dec

ades

, som

e so

urce

s rep

ort t

hat r

ecen

t eco

nom

ic tre

nds h

ave h

alted

and

may

reve

rse p

rogr

ess.

As o

verc

rowd

ing

is hi

ghes

t in

low-

inco

me h

ouse

hold

s it

is fre

quen

tly as

socia

ted

with

subs

tand

ard

hous

ing

and

othe

r asp

ects

of d

epriv

atio

n.

Polic

y im

plic

atio

ns: K

nowl

edge

of o

verc

rowd

ing

levels

and

the c

apac

ity to

link

this

info

rmat

ion

to h

ouse

hold

inco

me,

tenu

re an

d to

a va

riety

of o

ther

socio

dem

ogra

phic

varia

bles

can

help

to in

form

and

eva

luat

e ho

usin

g, ec

onom

ic an

d so

cial p

olici

es to

impr

ove

both

hea

lth a

nd e

quity

. The

re is

wid

e ac

cept

ance

of t

he n

eed

to in

crea

se th

e av

ailab

ility

of g

ood-

quali

ty

affo

rdab

le ho

usin

g fo

r all,

not l

east

the p

oore

st, m

ost s

ocial

ly ex

clude

d. H

owev

er, it

is cl

ear f

rom

the d

ata t

hat p

olici

es th

at ta

ckle

over

crow

ding

in th

e soc

ial re

nted

sect

or (a

nd in

the t

hree

lo

west

inco

me q

uint

iles,

irres

pect

ive o

f ten

ure g

roup

ing)

hav

e the

pot

entia

l to

impa

ct p

ositi

vely

on th

e live

s of t

he la

rges

t num

ber o

f peo

ple.

Prev

alen

ce o

f ove

rcro

wde

d ho

mes

by

tenu

re a

nd h

ouse

hold

inco

me

quin

tile

01234567

Soci

al re

nter

sPr

ivat

e re

nter

sO

wne

rs

Percentage of households

Qui

ntile

1 (l

owes

t inc

ome)

Qui

ntile

2Q

uint

ile 3

Qui

ntile

4Q

uint

ile 5

(hig

hest

inco

me)

Ana

lysis

and

inte

rpre

tatio

n: T

he ch

art a

bove

was

crea

ted

usin

g da

ta fr

om 2

8 90

1 va

lid ca

ses w

ithin

a

tota

l sur

vey

sam

ple

of 3

1 23

3 ho

useh

olds

(92

%).

Own

er-o

ccup

atio

n is

the

dom

inan

t te

nure

, re

pres

entin

g ap

prox

imat

ely 7

1% o

f the

sam

ple,

with

socia

l ren

ted

hom

es m

akin

g up

app

roxim

ately

19

%, a

nd t

he r

emain

ing

10%

priv

ate

rent

ed. H

owev

er, t

his

patte

rn o

f te

nure

dist

ribut

ion

is no

t re

flect

ed u

nifo

rmly

with

in e

ach

inco

me

quin

tile.

The

like

lihoo

d of

a h

ouse

hold

bein

g ov

ercr

owde

d in

crea

ses w

ith d

eclin

ing

hous

ehol

d in

com

e, as

illu

strat

ed b

y th

e fa

ct th

at th

e ov

erall

per

cent

age

of

over

crow

ded

hom

es in

the

lowe

st in

com

e qu

intil

e (3

.4%

) is m

ore

than

10

times

that

in th

e hi

ghes

t in

com

e qui

ntile

(0.3

%).

The

cha

rt sh

ows t

he p

erce

ntag

e of

ove

rcro

wdin

g in

eac

h te

nure

gro

up fo

r eac

h of

the

five

inco

me

quin

tiles

. Ove

rcro

wdin

g is e

viden

t in

each

tenu

re gr

oup

and

in ea

ch q

uint

ile. T

he h

ighe

st pe

rcen

tage

s of

ove

rcro

wdin

g fo

r all

inco

me

quin

tiles

are

foun

d in

the

socia

l ren

ted

sect

or w

here

as th

e lo

west

perc

enta

ges

are

foun

d am

ong

owne

r-oc

cupi

ers.

Qui

ntile

s 1–

3 ex

hibi

t no

t di

ssim

ilar

levels

of

over

crow

ding

in ea

ch te

nure

cate

gory

, impl

ying

that

ove

rcro

wdin

g is

not a

pro

blem

restr

icted

to th

e po

ores

t in

socie

ty.

Ove

rcro

wdi

ng b

y te

nure

gro

up a

nd h

ouse

hold

inco

me

Gre

at b

rita

in


KE

Y M

ESSA

GE:

Low

-inco

me

grou

ps a

nd s

ingl

e-pa

rent

hou

seho

lds

in N

orwa

y ar

e m

ore l

ikely

to re

port

dam

pnes

s in

their

dwe

lling

.

Dat

a sou

rce:

Dat

a are

take

n fro

m E

U-S

ILC

and

the n

atio

nal S

urve

y of

Livi

ng C

ondi

tions

(SLC

) by

Sta

tistic

s Nor

way.

Dat

a are

avail

able

at: h

ttp://

epp.

euro

stat.e

c.eur

opa.e

u an

d ht

tp://

www.

ssb.

no.

Dat

a des

crip

tion:

EU

-SIL

C: t

he p

aram

eter

is d

efin

ed a

s pro

porti

on o

f the

pop

ulat

ion

“livin

g in

a

dwell

ing

with

a le

akin

g ro

of, d

amp

walls

, flo

ors o

r fou

ndat

ions

, or r

ot in

win

dow

fram

es o

r flo

or”

(var

iable

HH

040)

. Pos

sible

valu

es ar

e “ye

s” or

“no”

. SLC

: the

par

amet

er is

def

ined

as “a

dwe

lling

that

is

dam

aged

by r

ot an

d m

ould

in so

me o

r all

the r

oom

s”. P

ossib

le va

lues

are “

yes”

or “n

o”.

Dat

a cov

erag

e: T

he d

ata

have

nat

iona

l cov

erag

e. EU

-SIL

C: d

ata

are

avail

able

annu

ally

from

200

4 to

200

9. SL

C: t

his a

nnua

l sur

vey c

hang

es to

pics

dur

ing

a thr

ee-y

ear c

ycle.

Hou

sing

cond

ition

s wer

e to

pics

in 1

997,

2001

, 200

4 an

d 20

07.

Soci

al d

imen

sion:

Thi

s ine

quali

ty fa

ct sh

eet s

tratif

ies d

ata

by a

ge, i

ncom

e (b

elow

or a

bove

60%

of

the m

edian

equi

valen

t, or

relat

ive p

over

ty le

vel)

and

hous

ehol

d ty

pe.

Con

text

: D

ampn

ess

has

been

ass

ociat

ed w

ith r

espi

rato

ry s

ympt

oms

in s

ever

al ep

idem

iolo

gica

l stu

dies

: a d

amp

envir

onm

ent

may

cau

se in

crea

sed

levels

of

bact

eria,

mou

lds

and

myc

otox

ins.

In

parti

cular

, ass

ociat

ions

bet

ween

mou

ld g

rowt

h an

d re

spira

tory

sym

ptom

s ha

ve b

een

exte

nsive

ly

studi

ed. I

n oc

cupa

tiona

l set

tings

, whe

re th

e exp

osur

e lev

els m

ay b

e ver

y hig

h, an

asso

ciatio

n be

twee

n m

ould

and

adve

rse h

ealth

effe

cts s

uch

as as

thm

a and

aller

gies

has

bee

n fo

und.

In p

rivat

e hou

seho

lds,

howe

ver, e

xpos

ure l

evels

are m

uch

lowe

r and

stud

ies h

ave f

ailed

to sh

ow su

ch as

socia

tions

.

Com

pare

d to

the

coun

tries

in th

e EU

, Nor

way

has

few

dwell

ings

with

repo

rted

dam

pnes

s (E

U:

15.9

%; N

orwa

y: 8.

2% (2

009)

). H

owev

er, N

orwa

y is c

ompa

rabl

e to

the o

ther

Sca

ndin

avian

coun

tries

(S

wede

n: 6

.6%

; Den

mar

k: 7

.8%

). Ten

ure s

tatu

s diff

ers b

etwe

en th

e hou

seho

ld ty

pes “

two

adul

ts wi

th

child

ren”

and

“sing

le pa

rent

”, in

that

coup

les w

ith ch

ildre

n ar

e mor

e lik

ely to

own

their

flat

or h

ouse

, wh

erea

s sin

gle p

aren

ts ar

e mor

e lik

ely to

live

in a

coop

erat

ive sh

areh

oldi

ng fl

at. A

hig

her p

erce

ntag

e of

sing

le pa

rent

s are

also

tena

nts.

Polic

y im

plic

atio

ns: T

he d

ata s

how

a soc

ial in

equa

lity w

ith re

gard

to d

ampn

ess i

n dw

ellin

gs, i

n th

at th

e hig

hest

frequ

ency

of r

epor

ted

dam

pnes

s is a

mon

g lo

w-in

com

e hou

seho

lds.

In

addi

tion,

ther

e are

diff

eren

ces b

etwe

en h

ouse

hold

type

s, wi

th m

ore s

ingl

e par

ents

than

sing

le pe

ople

or co

uples

with

child

ren

repo

rting

dam

pnes

s. In

Nor

way,

indo

or m

ainte

nanc

e is t

he

dwell

ing

owne

r’s re

spon

sibili

ty. P

eopl

e livi

ng in

free

hold

or c

oope

rativ

e sha

reho

ldin

g fla

ts th

us h

ave e

qual

oppo

rtuni

ties t

o de

al wi

th d

ampn

ess,

but p

eopl

e with

low

inco

me l

evels

may

no

t be a

ble t

o af

ford

the n

eces

sary

main

tena

nce w

ork.

The

hou

sing

mar

ket i

n N

orwa

y is u

nder

hig

h pr

essu

re; n

ot al

l pop

ulat

ion

grou

ps ar

e abl

e to

buy t

heir

own

flats

and

som

e the

refo

re

depe

nd o

n th

e mar

ket f

or re

ntal

flats.

If th

ere a

re d

ampn

ess i

ssue

s in

the r

ente

d fla

t and

the o

wner

s are

not

will

ing

to u

nder

take

main

tena

nce t

o im

prov

e the

indo

or en

viron

men

t, th

ese

resid

ents

have

few

oppo

rtuni

ties t

o ch

ange

their

hou

sing

cond

ition

s. A

n ob

ligat

ory

surv

ey re

port

on re

ntal

flats

coul

d be

a re

gulat

ory

step

to se

cure

a d

ecen

t ind

oor e

nviro

nmen

t. In

ad

ditio

n, a

syste

m w

here

by lo

w-in

com

e gro

ups c

ould

be g

iven

a fav

oura

ble l

oan

to b

e abl

e to

buy t

heir

own

flat c

ould

also

be c

onsid

ered

. Sim

ilarly

, loan

s cou

ld b

e give

n to

hom

e-ow

ners

who

are u

nabl

e to

perfo

rm th

e nec

essa

ry m

ainte

nanc

e for

econ

omic

reas

ons.

Sing

le (1

6–24

yea

rs)

5Si

ngle

(25–

44 y

ears

)5

Sing

le (4

5–66

yea

rs)

3Si

ngle

(>66

yea

rs)

2Tw

o ad

ults

(16–

44

year

s)5

Two

adul

ts (4

5–66

ye

ars)

3

Two

adul

ts (>

66 y

ears

)1

Two

adul

ts w

ith c

hild

ren

aged

0–6

yea

rs3

Two

adul

ts w

ith c

hild

ren

aged

7–1

9 ye

ars

4

Sing

le p

aren

ts7

Vita

li: th

is is

a ta

ble

incl

uded

in th

e fa

ct s

heet

Rep

orte

d da

mpn

ess

by h

ouse

hold

type

(SLC

, 200

7)Se

lf-re

port

ed d

ampn

ess

by re

lativ

e po

vert

y (E

U-S

ILC

, 200

9)

02468101214161820

2004

2005

2006

2007

2008

2009

Year


Hou

seho

lds

abov

e re

lativ

e po

verty

leve

lH

ouse

hold

s be

low

rela

tive

pove

rty le

vel

Hou

seho

ld t

ype

Rep

orte

dda

mpn

ess

(%)

Ana

lysis

and

inte

rpre

tatio

n: T

he n

umbe

r of r

epor

ts of

dam

pnes

s is l

owes

t in

the h

ighe

st ag

e gro

up.

The

hou

seho

ld t

ype

that

mos

t fre

quen

tly r

epor

ts da

mpn

ess

cons

ists

of o

ne a

dult

with

chi

ldre

n.

The

re is

mor

e rep

orte

d da

mpn

ess i

n th

e gro

up ea

rnin

g be

low

the r

elativ

e pov

erty

leve

l (in

the p

erio

d 20

04–2

009,

arou

nd 9

–10%

of t

he to

tal p

opul

atio

n ha

d an

inco

me

level

belo

w th

e re

lative

pov

erty

lev

el). S

ingl

e-pa

rent

fam

ilies

are

a m

ajor s

ubse

t with

in th

is gr

oup,

cons

titut

ing

arou

nd 1

7–23

%. I

n co

mpa

rison

, the

figu

re fo

r hou

seho

lds w

ith tw

o ad

ults

with

child

ren

is ar

ound

8%

.

Repo

rting

dam

pnes

s is s

ubjec

tive a

nd d

oes n

ot gi

ve an

indi

catio

n of

the e

xten

siven

ess o

f the

pro

blem

; th

us it

may

not

repr

esen

t a se

rious

hou

sing

prob

lem. H

owev

er, th

e psy

chol

ogica

l fac

tor i

s im

porta

nt,

since

peo

ple

may

be

afra

id o

f dev

elopi

ng a

ssoc

iated

hea

lth p

robl

ems,

or m

ay re

late

alrea

dy e

xistin

g he

alth

effe

cts t

o da

mpn

ess i

n th

e dw

ellin

g. In

add

ition

, sin

gle

pare

nts c

onsti

tute

a la

rge

prop

ortio

n of

the

low-

inco

me

grou

p an

d ar

e m

ore

ofte

n te

nant

s tha

n co

uples

with

chi

ldre

n. B

oth

low

inco

me

and

tenu

re st

atus

may

influ

ence

the a

bilit

y to

deal

with

the d

ampn

ess p

robl

em o

r to

mov

e to

a drie

r dw

ellin

g.

dam

pnes

s in

dw

ellin

g by

age

, inc

ome

and

hous

ehol

d ty

peN

orw

ay


KE

Y M

ESSA

GE:

Poor

ly ed

ucat

ed an

d ove

rcro

wded

hous

ehol

ds in

Serb

ia an

d Mon

tene

gro

are s

uffe

ring

from

a lac

k of

hom

e hea

ting,

which

also

affe

cts t

heir

healt

h sta

tus.

Dat

a so

urce

: Dat

a ar

e ta

ken

from

the

study

“Stu

ck in

the

Past

” car

ried

out b

y U

ND

P in

200

4 in

Se

rbia

and

Mon

tene

gro.

Furth

er d

etail

s can

be a

cces

sed

from

the m

ain re

port

at: h

ttp://

www.

undp

.or

g/en

ergy

/doc

s/Stu

ck_i

n_th

e_Pa

st.pd

f.

Dat

a des

crip

tion:

The

par

amet

er “h

ome h

eatin

g” is

pre

sent

ed as

“squ

are m

etre

s of h

eate

d sp

ace p

er

hous

ehol

d m

embe

r”. P

aram

eter

s wer

e def

ined

thro

ugh

two

surv

eys f

orm

ing p

art o

f the

UN

DP

study

wh

ich ex

amin

ed h

ouse

hold

ener

gy u

sage

.

Dat

a cov

erag

e: T

he d

ata u

sed

are f

rom

200

2–20

03. T

he su

rvey

was

repe

ated

in 2

008

but t

he m

ain

resu

lts (u

npub

lishe

d) w

ere v

ery s

imila

r to

thos

e fro

m 2

002–

2003

and

indi

cate

no

signi

fican

t cha

nges

in

relat

ion

to th

e ide

ntifi

ed in

equa

lities

.

Soci

al d

imen

sion:

Thi

s ine

quali

ty fa

ct sh

eet s

tratif

ies th

e am

ount

of h

eate

d sp

ace

per h

ouse

hold

m

embe

r by

lev

el of

edu

catio

n (p

rimar

y sc

hool

and

uni

versi

ty/c

olleg

e) a

nd h

ouse

hold

siz

e an

d co

mpo

sitio

n (la

rge

hous

ehol

ds,

hous

ehol

ds w

ith t

wo o

r m

ore

child

ren

and

thre

e-ge

nera

tion

hous

ehol

ds),

and

puts

this

into

con

text

with

var

iatio

ns in

hea

lth s

tatu

s (p

ropo

rtion

of h

ouse

hold

m

embe

rs wi

thou

t hea

lth p

robl

ems).

Con

text

: Fac

tors

such

old

hou

sing,

poor

insu

latio

n, h

eatin

g with

“dirt

y fue

ls” (w

ood

and

ligni

te co

al),

indo

or p

ollu

tion

and

over

crow

ding

are a

ssoc

iated

with

pov

erty.

Poo

r hou

seho

lds p

ract

ise ri

sky f

orm

s of

ener

gy sa

ving

and

ofte

n he

at o

nly a

bout

half

their

livin

g sp

ace i

n or

der t

o sa

ve o

n en

ergy

costs

.

As a

resu

lt, th

e re

lative

am

ount

of h

eate

d sp

ace

strat

ifies

by

inco

me

per h

ouse

hold

mem

ber a

nd is

as

socia

ted

with

a ra

nge o

f soc

ial o

r hou

seho

ld-r

elate

d ch

arac

teris

tics.

The

pro

blem

s ind

icate

d by

the

data

are

less

mar

ked

in u

rban

are

as b

ut re

pres

ent t

he n

orm

al liv

ing

cond

ition

s fo

r hou

seho

lds

with

low

inco

me

in r

ural

area

s. T

his

type

of h

ouse

hold

typi

cally

has

a

subs

tand

ard

hous

e with

low

effic

iency

hea

ting

devic

es an

d an

amou

nt o

f hea

ted

spac

e per

hou

seho

ld

mem

ber t

hat i

s belo

w 8

m2 ;

thes

e fac

tors

are a

lso cl

osely

asso

ciate

d wi

th h

ealth

-rela

ted

ineq

ualit

ies.

Polic

y im

plic

atio

ns: T

here

is a

ver

y str

ong

corre

latio

n be

twee

n po

verty

and

lack

of h

eatin

g sp

ace.

Redu

cing

the

heat

ed a

rea

of a

hou

seho

ld c

an le

ad to

tens

ions

and

dys

func

tiona

l re

latio

nshi

ps b

etwe

en h

ouse

hold

mem

bers.

Insu

fficie

nt h

eate

d sp

ace c

an h

ave c

onse

quen

ces f

or th

e hea

lth an

d qu

ality

of l

ife o

f the

hou

seho

ld. C

opin

g str

ateg

ies ar

e par

ticul

arly

limite

d fo

r poo

r hou

seho

lds,

whos

e lac

k of

capi

tal o

ften

forc

es th

em to

adop

t dan

gero

us an

d in

effic

ient e

nerg

y-sa

ving

strat

egies

.

Bette

r gov

erna

nce a

nd a

mor

e com

preh

ensiv

e pol

icy ar

e nee

ded.

Polic

y-m

aker

s nee

d to

esta

blish

an en

forc

eabl

e con

cept

of p

rope

rty ri

ghts

and

publ

ic go

ods,

to b

uild

hou

sing

capa

city,

to

impr

ove i

nstit

utio

nal a

nd co

rpor

ative

gov

erna

nce,

to d

evelo

p be

tter p

olice

s and

refo

rms,

and

impr

ove i

nfor

mat

ion

flows

and

struc

ture

s.

Hea

ted

dwel

ling

spac

e by

hou

seho

ld s

ocia

l ind

icat

ors

010

2030

4050

6070

80

Prim

ary

scho

ol

Uni

vers

ity/c

olle

ge

Hou

seho

ld w

ith fi

veor

mor

e m

embe

rs

Hou

seho

ld w

ith tw

oor

mor

e ch

ildre

n

Thre

e-ge

nera

tion

hous

ehol

d

No

repo

rted

heal

th p

robl

ems

Educationalstatus

Householdtype

Healthstatus

Perc

enta

ge o

f hou

seho

lds

Mor

e th

an 2

0 m

² of

heat

ed s

pace

per

hous

ehol

d m

embe

r

Less

than

10

m² o

fhe

ated

spa

ce p

erho

useh

old

mem

ber

Ana

lysis

and

inte

rpre

tatio

n: P

oor h

ouse

hold

s he

at o

nly

abou

t half

their

livin

g sp

ace

in o

rder

to

save

on

ener

gy co

sts. O

n av

erag

e mor

e tha

n on

e in

four

hou

seho

lds (

26%

) hea

t les

s tha

n 10

m2 p

er

perso

n, c

onsid

ered

the

nece

ssar

y m

inim

um. H

owev

er, th

is pr

opor

tion

is in

crea

sed

for h

ouse

hold

s wi

th lo

w ed

ucat

ion

level

and

for l

arge

hou

seho

lds.

Lack

of h

eat,

over

crow

ding

and

indo

or p

ollu

tion

com

e tog

ethe

r as i

nequ

alitie

s in

poor

hou

seho

lds a

nd re

sult

in se

vere

redu

ctio

ns in

usa

ble f

loor

spac

e in

win

ter t

ime (

beca

use o

f a re

strict

ed h

eate

d ar

ea),

which

disr

upts

the n

orm

al ho

useh

old

dyna

mics

an

d pu

ts pr

essu

re o

n pe

rsona

l and

hyg

iene h

abits

as w

ell as

socia

l and

cultu

ral n

orm

s.

The

soc

ial a

nd h

ouse

hold

-rela

ted

diffe

renc

es a

re a

ssoc

iated

with

hea

lth-r

elate

d in

equa

lities

, in

dica

ting

that

in h

ouse

hold

s with

less

hea

ted

spac

e, th

ere i

s an

incr

ease

d pr

obab

ility

of h

ouse

hold

s re

porti

ng h

ealth

pro

blem

s. T

he d

iffer

ence

s are

eve

n m

ore

signi

fican

t for

hea

ted

spac

e pe

r mem

ber

of le

ss th

an 5

m2 .

lack

of h

ome

heat

ing

by e

duca

tion

and

hou

seho

ld c

ompo

siti

onSe

rbia

and

mon

tene

gro


KE

Y M

ESSA

GE:

Male

wor

kers

in C

roat

ia ar

e m

ore

likely

to

suffe

r an

occ

upat

iona

l in

jury

than

fem

ale w

orke

rs. M

ost w

ork

accid

ents

occu

r in

the

agric

ultu

re, f

ores

try a

nd

fishi

ng, m

anuf

actu

ring,

wate

r sup

ply a

nd w

aste

man

agem

ent,

and

cons

truct

ion

sect

ors.

Dat

a sou

rce:

Dat

a are

take

n fro

m th

e rep

ort o

n acc

iden

t at w

ork n

otifi

catio

ns by

the C

roat

ian N

atio

nal

Insti

tute

of P

ublic

Hea

lth. D

ata f

rom

200

8 on

ward

s, br

oken

dow

n by

sex,

can

be d

ownl

oade

d fro

m

the I

nstit

ute’s

web

site

at: h

ttp://

www.

hzjz.

hr/p

ublik

acije

.htm

.

Dat

a de

scrip

tion:

Dat

a co

ver

all r

ecog

nize

d wo

rk a

ccid

ents

which

res

ult

in a

bsen

ce fr

om w

ork.

Wor

k ac

ciden

ts ar

e ac

ciden

ts su

ffere

d by

insu

red

peop

le (in

sura

nts)

in c

onne

ctio

n wi

th th

eir w

ork,

or w

hilst

trav

ellin

g on

com

pany

bus

ines

s. A

ccor

ding

to C

roat

ian le

gisla

tion,

com

mut

ing

accid

ents

(def

ined

as tr

avel

to an

d fro

m w

ork)

are a

lso in

clude

d.

Dat

a co

vera

ge: D

ata

cove

r all

fata

l and

non

-fat

al wo

rk a

ccid

ents,

inclu

ding

com

mut

ing

accid

ents,

br

oken

dow

n by

econ

omic

sect

or an

d se

x.

Soci

al d

imen

sion:

Alth

ough

wor

k ac

ciden

ts co

ntrib

ute

to n

o m

ore

than

3–6

% o

f tot

al sic

k lea

ve,

their

shar

e is s

igni

fican

t, gi

ven

the c

onse

quen

ces t

hey c

ause

. Whi

le ca

ses o

f sick

ness

caus

e an

aver

age

loss

of 2

0 to

29

days

, a w

ork

accid

ent l

eads

to an

aver

age o

f ove

r 60

days

of a

bsen

ce fr

om w

ork,

and

1–5%

of i

njur

ies en

d in

per

man

ent d

isabi

lity.

Con

text

: On

31 D

ecem

ber 2

009,

Cro

atia

had

1 53

0 23

3 ac

tive

insu

rant

s, wh

ich co

nstit

utes

41.

3%

of th

e wo

rkin

g-ag

e po

pulat

ion

(age

s 15–

65),

or 8

6.7%

of t

he a

ctive

labo

ur fo

rce

(the

empl

oym

ent

male

to fe

male

ratio

was

54:

46).

Cro

atia

annu

ally r

egist

ers o

ver 2

0 00

0 wo

rk-r

elate

d in

jurie

s. In

oth

er

word

s, be

twee

n 15

and

17 em

ploy

ees p

er 1

000

are a

nnua

lly in

jure

d at

wor

k.

In 2

009

the

over

all ra

te w

as 1

2.8

per 1

000

empl

oyee

s (9.

8/10

00 e

xclu

ding

com

mut

ing

accid

ents)

. T

he a

vera

ge r

ate

was

16.7

/100

0 fo

r m

ales,

and

7.1/

1000

for

fem

ales.

By i

ndus

try s

ecto

r, th

e hi

ghes

t spe

cific

rate

s wer

e in

agr

icultu

re, f

ores

try a

nd fi

shin

g (2

4.1/

1000

), wa

ter s

uppl

y an

d wa

ste

man

agem

ent (

20.8

/100

0), m

anuf

actu

ring

(20.

2/10

00) a

nd co

nstru

ctio

n (1

9.5/

1000

). T

hese

figu

res,

which

plac

e C

roat

ia am

ong

the

coun

tries

with

a lo

w ra

te o

f reg

ister

ed w

ork

inju

ries,

have

bee

n re

lative

ly co

nsta

nt o

ver t

he la

st de

cade

.

Polic

y im

plic

atio

ns: O

n 29

Dec

embe

r 20

08, t

he C

roat

ian G

over

nmen

t ado

pted

the

Nat

iona

l Occ

upat

iona

l Hea

lth a

nd S

afet

y Pr

ogra

mm

e 20

09–2

013.

The

stra

tegi

c aim

of t

he

Prog

ram

me i

s to

ensu

re sa

fe an

d he

althy

wor

k tra

vel a

nd w

orkp

lace c

ondi

tions

. To

achi

eve t

his a

im, a

dopt

ion

of th

e EU

Occ

upat

iona

l Hea

lth an

d Sa

fety

at W

ork

Stra

tegy

200

7–20

12

is be

ing

cons

ider

ed. I

mpl

emen

tatio

n of

a na

tiona

l pro

gram

me w

ill h

elp to

solve

a m

ultit

ude o

f pro

blem

s in

the a

rea o

f occ

upat

iona

l hea

lth an

d sa

fety,

such

as il

l-sui

ted

educ

atio

n, la

ck o

f ha

rmon

izat

ion

and

rese

arch

pro

jects,

uns

yste

mat

ic im

plem

enta

tion

and,

mos

t of a

ll, ne

glig

ence

of o

ccup

atio

nal i

njur

y and

dise

ase p

reve

ntio

n m

easu

res.

Fund

amen

tal o

bjec

tives

of o

ccup

atio

nal h

ealth

care

and

safe

ty ar

e: re

duct

ion

of o

ccup

atio

nal i

njur

ies an

d di

seas

es, a

nd im

prov

emen

t of w

orke

rs’ h

ealth

stat

us b

y mea

ns o

f pre

vent

ion

and

redu

ctio

n of

econ

omic

loss

es d

ue to

occ

upat

iona

l inj

uries

and

dise

ases

(sick

leav

e, pr

emat

ure a

nd d

isabi

lity p

ensio

ns).

Wor

k-re

late

d in

jury

rate

by

sex

and

econ

omic

sec

tor (

2009

)

05101520253035

Hea

lth a

ndso

cial

wor

kAc

com

mo-

datio

n an

d fo

odse

rvic

e

Tran

spor

tatio

nan

d st

orag

eC

onst

ruct

ion

Man

ufac

turin

gW

ater

su

pply

and

was

tem

anag

emen

t

Agric

ultu

re,

fore

stry

and

fishi

ng

Injury rate/1000 employees

Mal

esFe

mal

esTo

tal

Ana

lysis

and

inte

rpre

tatio

n: In

200

9 th

e m

ajorit

y (7

6.2%

) of w

ork-

relat

ed in

jurie

s, fo

llowi

ng th

e tre

nd o

f pre

cedi

ng y

ears,

wer

e su

stain

ed a

t the

wor

kplac

e, wh

ile 2

3.8%

hap

pene

d on

the

way

to

or f

rom

wor

k. A

ppro

ximat

ely 7

0% o

f ca

ses

were

wor

k in

jurie

s to

male

s, 30

% t

o fe

male

s. M

ales

suffe

red

85.3

% o

f acc

iden

ts at

wor

k an

d on

ly 14

.7%

whi

le co

mm

utin

g, wh

ile fo

r fem

ales t

he ra

tio

was

58.8

%:4

1.2%

. The

majo

rity

of w

ork

accid

ents

susta

ined

by

male

s oc

curre

d in

agr

icultu

re,

fore

stry

and

fishi

ng (2

9.0/

1000

), m

anuf

actu

ring

(25.

7/10

00),

wate

r sup

ply

and

waste

man

agem

ent

(23.

6/10

00) a

nd c

onstr

uctio

n (2

1.8/

1000

); th

ose

susta

ined

by

fem

ales o

ccur

red

in a

ccom

mod

atio

n an

d fo

od se

rvice

activ

ities

(13.

4/10

00),

agric

ultu

re, fo

restr

y and

fish

ing (

11.1

/100

0), h

ealth

and

socia

l wo

rk (1

0.7/

1000

) and

man

ufac

turin

g (1

0.2/

1000

). M

en w

ere

mor

e ex

pose

d to

risk

: 94.

7% o

f fat

al ac

ciden

ts we

re su

ffere

d by

male

wor

kers

(5.7

/100

000

), on

ly 5.

3% b

y fem

ale w

orke

rs (0

.4/1

00 0

00).

Of t

hese

, 83%

of t

hose

susta

ined

by m

ales,

but n

one o

f tho

se su

stain

ed by

fem

ales,

were

wor

k-re

lated

. T

he m

ost d

ange

rous

sec

tor,

with

the

high

est n

umbe

r of

inju

ries

or fa

talit

ies a

t the

wor

kplac

e, is

agric

ultu

re, f

ores

try an

d fis

hery

, fol

lowe

d by

wat

er su

pply

and

waste

man

agem

ent.

wor

k-re

late

d in

juri

es b

y se

x an

d ec

onom

ic s

ecto

rC

roat

ia


KE

Y M

ESSA

GE:

The

vast

majo

rity o

f all

trans

port-

relat

ed d

eath

s in

Malt

a occ

ur am

ong

male

s, wi

th th

e hig

hest

prop

ortio

n in

the 1

5–29

year

age g

roup

. Ove

r the

last

deca

de, t

he

prop

ortio

n of

fem

ale d

eath

s has

incr

ease

d.

Dat

a sou

rce:

Dat

a are

take

n fro

m th

e Nat

iona

l Mor

talit

y Reg

ister

held

by t

he D

irect

orat

e of H

ealth

In

form

atio

n an

d Re

sear

ch. T

he re

giste

r hol

ds m

orta

lity d

ata f

or M

altes

e res

iden

ts.

Dat

a de

scrip

tion:

The

par

amet

er is

def

ined

as “

trans

port

accid

ents”

acc

ordi

ng to

IC

D-1

0 co

des

V01

–V99

.

Dat

a cov

erag

e: A

nnua

l dat

a are

avail

able.

Dat

a pre

sent

ed ar

e for

200

1–20

10.

Soci

al d

imen

sion:

Thi

s ine

quali

ty fa

ct sh

eet s

tratif

ies “t

rans

port-

relat

ed m

orta

lity”

by

targ

et g

roup

(a

dult

male

, adu

lt fe

male

and

chi

ld)

acco

rdin

g to

the

type

of t

rans

port

accid

ent.

The

num

ber

of

trans

port-

relat

ed d

eath

s for

this

perio

d wa

s 168

. Sin

ce th

e num

bers

are l

ow, d

ata m

ust b

e int

erpr

eted

wi

th ca

utio

n.

Con

text

: Malt

a is

a de

nsely

pop

ulat

ed c

ount

ry w

ith 1

307

peop

le pe

r km

2 . Si

nce

the

1980

s th

e nu

mbe

r of l

icens

ed v

ehicl

es p

er c

apita

has

rise

n ste

adily

, fro

m 0

.33

in 1

986

to 0

.63

in 2

001.

At t

he

end

of 2

009,

the n

umbe

r was

297

776

or 0

.72

vehi

cles p

er ca

pita

.

Tran

spor

t ac

ciden

ts ac

coun

ted

for

10.4

% o

f all

dea

ths

in t

he 1

5–44

yea

r ag

e gr

oup

in 2

008,

repr

esen

ting 4

.4%

of a

ll th

e pot

entia

l yea

rs of

life

lost

in th

ose u

nder

the a

ge o

f 65.

Ove

r the

refe

renc

e pe

riod,

drivi

ng li

cenc

es in

Malt

a we

re h

eld p

redo

min

antly

by

male

s (61

% o

f lice

nce

hold

ers).

The

pr

opor

tions

of m

ale an

d fe

male

lice

nce h

olde

rs in

the y

oung

er ag

e gro

ups t

end

to b

e clo

ser, i

mpl

ying

th

at m

ore

youn

ger f

emale

s are

taki

ng u

p dr

iving

. Nev

erth

eless

, a g

reat

er p

erce

ntag

e of

male

s tha

n fe

male

s tak

e up

driv

ing

at a

n ea

rly a

ge. I

n 20

09 5

9% o

f all

drivi

ng li

cenc

es in

the

18–2

4 ye

ar a

ge

grou

p we

re h

eld b

y male

s. A

t pre

sent

93%

of m

otor

cycle

lice

nce h

olde

rs ar

e male

.

Ove

r the

last

deca

de, m

easu

res h

ave b

een

take

n to

slow

dow

n tra

ffic i

n vil

lages

, suc

h as

cons

truct

ing

mor

e spe

ed b

umps

and

insta

lling

spee

d ca

mer

as to

det

er sp

eedi

ng. A

pub

lic tr

ansp

ort r

efor

m p

olicy

wa

s lau

nche

d in

July

2011

, with

the

aim o

f mak

ing

publ

ic tra

nspo

rt m

ore

effic

ient,

mor

e ef

fect

ive

and

oper

atin

g ov

er a

wide

r net

work

.

Polic

y im

plic

atio

ns: S

ince

the v

ast m

ajorit

y of

tran

spor

t-re

lated

dea

ths o

ccur

amon

g m

ales,

any

polic

y th

at m

anag

es to

redu

ce tr

ansp

ort-

relat

ed m

orta

lity

woul

d au

tom

atica

lly re

duce

se

x-re

lated

ineq

ualit

y. En

cour

agin

g pu

blic

trans

port

use a

nd al

tern

ative

mod

es o

f tra

nspo

rt am

ong

youn

ger a

ge g

roup

s, es

pecia

lly m

ales,

well

befo

re th

ey ta

ke u

p dr

iving

may

redu

ce th

e nu

mbe

r of d

eath

s occ

urrin

g in

the 1

5–29

year

age g

roup

. In

addi

tion,

furth

er en

forc

emen

t and

augm

enta

tion

of p

enalt

ies fo

r driv

ing

dang

erou

sly (s

uch

as sp

eedi

ng) m

ay b

e call

ed fo

r to

dete

r driv

ers f

rom

spee

ding

and

from

drin

king

and

drivi

ng. B

reat

haly

ser t

estin

g of

all c

ar d

river

s inv

olve

d in

a tra

ffic a

ccid

ent m

ay w

arra

nt fu

rther

disc

ussio

n. In

par

allel,

the f

act t

hat t

he

popu

latio

n is

incr

easin

gly a

gein

g m

ust b

e not

ed. C

ar-f

ree z

ones

will

def

inite

ly m

ake a

reas

safe

r for

ped

estri

ans.

Mor

talit

y ra

te b

y ty

pe o

f roa

d us

er (a

vera

ge 2

001–

2010

)

051015202530

Pede

stria

nPe

dal

cycl

ist

Mot

orcy

clis

tC

aroc

cupa

ntPi

ck-u

ptru

ck/v

anoc

cupa

nt

Hea

vy g

oods

vehi

cle

occu

pant

Bus

occu

pant


Adul

t mal

es

Adul

t fem

ales

Chi

ldre

n (0

–14

year

s)

Ana

lysis

and

inte

rpre

tatio

n: B

etwe

en 2

001

and

2010

, 81%

of a

ll tra

nspo

rt-re

lated

dea

ths o

ccur

red

amon

g m

ales a

nd 1

9% a

mon

g fe

male

s. T

he a

vera

ge m

ale to

fem

ale m

orta

lity

ratio

dec

reas

ed o

ver

the

year

s, fro

m 5

:1 in

the

perio

d 20

01–2

005

to 3

.6:1

in th

e pe

riod

2006

–201

0. T

he lo

west

ratio

, 2.

2:1,

was i

n 20

10. T

he la

rges

t num

ber o

f male

dea

ths o

ccur

red

in th

e 15

–29

year

age

gro

up, w

hile

mor

talit

y in

fem

ales w

as h

ighe

st in

the 6

0+ ye

ar ag

e gro

up. 3

4% o

f all

male

dea

ths o

ccur

red

amon

g ca

r occ

upan

ts (7

3% d

river

s and

27%

pas

seng

ers)

and

30%

amon

g m

otor

cycle

use

rs. 5

0% o

f all

fem

ale

deat

hs o

ccur

red

amon

g pe

destr

ians a

nd 3

4% am

ong

car o

ccup

ants.

The

re w

ere n

o tra

nspo

rt-re

lated

de

aths

relat

ed to

mot

orcy

cle u

se am

ong

fem

ales o

ver t

he 1

0-ye

ar p

erio

d. It

is im

porta

nt to

not

e tha

t m

ost d

eath

s occ

urre

d on

Sat

urda

ys an

d Su

nday

s.

The

disc

repa

ncy

betw

een

prop

ortio

ns o

f male

and

fem

ale li

cenc

e ho

lder

s may

exp

lain

part

of th

e di

ffere

nce b

etwe

en th

e tot

al nu

mbe

rs of

tran

spor

t-re

lated

dea

ths b

y sex

, not

taki

ng in

to ac

coun

t the

po

ssib

ility

that

a gr

eate

r pro

porti

on o

f fem

ale li

cenc

e hol

ders

do n

ot ac

tuall

y driv

e. In

addi

tion,

male

s te

nd to

get

their

driv

ing

licen

ce e

arlie

r, po

ssib

ly ac

coun

ting

for t

he b

igge

r pro

porti

on o

f dea

ths i

n m

ales i

n th

e you

nger

age g

roup

s. M

ales m

ay al

so b

e driv

ing

mor

e dan

gero

usly

than

fem

ales.

A la

rger

pr

opor

tion

of m

ales t

ake u

p m

otor

cycle

driv

ing,

which

in th

is re

fere

nce p

erio

d ac

coun

ts fo

r 41

deat

hs

or 3

0% o

f all

male

tran

spor

t-re

lated

dea

ths.

Sinc

e m

ost d

eath

s occ

ur a

t wee

kend

s, so

me

trans

port

fata

lities

may

be

relat

ed to

drin

king

and

driv

ing.

It is

also

impo

rtant

to n

ote

the

num

ber o

f dea

ths

of u

npro

tect

ed o

r les

s pro

tect

ed ro

ad u

sers

such

as p

edes

trian

s, pe

dal c

yclis

ts an

d m

otor

cycle

use

rs.

Tran

spor

t-re

late

d m

orta

lity

by s

ex a

nd a

gem

alta


KE

Y M

ESSA

GE:

Mor

talit

y fro

m a

ccid

enta

l poi

soni

ng in

Pol

and

is alm

ost f

ive ti

mes

hi

gher

in m

ales t

han

in fe

male

s. T

here

is a

hig

h in

crea

se in

dea

th ra

tes i

n th

e 40–

64 ye

ar

age g

roup

. Alco

hol i

s the

main

caus

e of m

orta

lity f

rom

accid

enta

l poi

soni

ng.

Dat

a so

urce

: Acc

iden

tal p

oiso

ning

-rela

ted

mor

talit

y da

ta a

re ta

ken

from

dea

th c

ertif

icate

s, wi

th

caus

e of

dea

th b

ased

on

ICD

-10

code

s (a

dopt

ed in

199

7). D

ata

are

rout

inely

gat

here

d by

the

Po

lish

Min

istry

of H

ealth

. Dat

a by

age

gro

up a

nd se

x on

a n

atio

nal l

evel

with

urb

an a

nd ru

ral a

rea

strat

ifica

tion

are a

vaila

ble f

rom

the C

entra

l Sta

tistic

al O

ffice

at: h

ttp://

www

.stat

.gov

.pl/g

us.

Dat

a des

crip

tion:

The

indi

cato

r is d

efin

ed as

accid

enta

l poi

soni

ng m

orta

lity (

see I

CD

-10

defin

ition

of

“acc

iden

tal p

oiso

ning

by a

nd ex

posu

re to

nox

ious

subs

tanc

es”,

code

s X40

–X49

).

Dat

a co

vera

ge: T

he d

ata

have

nat

iona

l cov

erag

e at

the

mun

icipa

lity

level,

stra

tified

by

urba

n an

d ru

ral a

rea,

age

grou

p an

d se

x. D

ata

are

avail

able

in c

urre

nt IC

D-1

0 cla

ssifi

catio

n sin

ce 2

002.

Thi

s an

alysis

is b

ased

on

data

from

200

9.

Soci

al d

imen

sion:

Thi

s ine

quali

ty fa

ct sh

eet s

tratif

ies m

orta

lity

from

acc

iden

tal p

oiso

ning

by

age

and

sex.

The

re a

re fo

ur a

lmos

t equ

al ag

e gr

oups

. In

addi

tion,

dat

a ca

n be

bro

ken

down

by

urba

n or

ru

ral r

esid

ence

.

Con

text

: Bet

ween

150

0 an

d 20

00 d

eath

s as a

resu

lt of

accid

enta

l poi

soni

ng ar

e rec

orde

d in

Pol

and

ever

y ye

ar. A

ccid

enta

l poi

soni

ng is

mos

t ofte

n ca

used

by

subs

tanc

e ab

use,

reck

lessn

ess o

r sub

stanc

e gi

ven

or ta

ken

in er

ror o

r ina

dver

tent

ly.

The

re a

re d

iffer

ent

abso

rptio

n ro

utes

: via

the

dig

estiv

e tra

ct (

med

icine

s, alc

ohol

, mus

hroo

ms,

dete

rgen

ts, a

nd so

on)

, the

resp

irato

ry tr

act (

carb

on m

onox

ide,

for e

xam

ple)

, the

skin

and

muc

ous

mem

bran

es (a

tropi

ne, p

estic

ides

, cau

stic

subs

tanc

es, a

nd so

on)

as w

ell a

s the

soft

tissu

es (v

ario

us

drug

s).

The

mos

t com

mon

caus

es o

f mor

talit

y fro

m ac

ciden

tal p

oiso

ning

in P

olan

d ar

e alco

hol a

buse

, leak

s fro

m g

as in

stalla

tions

(main

ly in

old

bui

ldin

gs o

r dwe

lling

s inh

abite

d by

poo

r fam

ilies

), a l

ack

of sa

fe

stora

ge fo

r med

icine

s and

chem

icals

and

child

ren

lacki

ng ap

prop

riate

care

.

Polic

y im

plic

atio

ns: I

n Po

land,

abou

t 200

0 de

aths

are r

epor

ted

ever

y yea

r as a

resu

lt of

accid

enta

l poi

soni

ng. N

early

92%

of s

uch

deat

hs ar

e cau

sed

by al

coho

l con

sum

ptio

n an

d ex

posu

re

to g

ases

. Peo

ple a

ged

betw

een

40 an

d 64

year

s hav

e a h

ighe

r risk

of m

orta

lity f

rom

accid

enta

l poi

soni

ng th

an an

y oth

er ag

e gro

up.

Key

pol

icy-m

akin

g ac

tiviti

es sh

ould

be f

ocus

ed o

n na

tiona

l cam

paig

ns ag

ainst

alcoh

olism

and

drug

abus

e on

the o

ne h

and,

and

mod

erni

zatio

n of

old

gas

insta

llatio

ns an

d pr

even

tion

of

gas l

eaks

(esp

ecial

ly in

old

bui

ldin

gs) o

n th

e oth

er.

Acc

iden

tal p

oiso

ning

mor

talit

y ra

te b

y se

x, a

ge a

nd u

rban

/rura

l res

iden

ce

02468101214161820

laruR

nabrU

laruR

nabrU

Mal

es

Mortality rate/100 000 inhabitants

0–19

yea

rs20

–39

year

s40

–64

year

s65

+ ye

ars

Fem

ales

Ana

lysis

and

inte

rpre

tatio

n: T

here

is a

clear

diff

eren

ce b

etwe

en th

e sex

es in

accid

enta

l poi

soni

ng-

relat

ed m

orta

lity (

abou

t five

tim

es m

ore d

eath

s occ

ur am

ong

male

s tha

n fe

male

s) an

d th

ere a

re al

so

signi

fican

t disp

ariti

es be

twee

n ag

e gro

ups.

Mor

e det

ailed

dat

a sho

w th

e pea

k of a

ccid

enta

l poi

soni

ng-

relat

ed m

orta

lity a

mon

g th

e 40–

49 ye

ar ag

e gro

up in

bot

h se

xes.

A d

ecre

asin

g de

ath

rate

is o

bser

ved

afte

r the

age

of 5

0. In

male

s, hi

gher

acc

iden

tal p

oiso

ning

-rela

ted

deat

h ra

tes

occu

r in

rura

l are

as

(exc

ept a

mon

g th

ose

unde

r 20

yea

rs) w

hile

for

fem

ales,

high

er r

ates

are

foun

d in

urb

an s

ettin

gs

(exc

ept a

mon

g th

ose o

ver 6

4 ye

ars).

Acc

iden

tal p

oiso

ning

-rela

ted

mor

talit

y pe

aked

in P

olan

d in

200

5–20

06, w

hen

a hi

gh n

umbe

r of

de

aths

was

reco

rded

. Sin

ce th

en, t

here

has

bee

n a

decli

ning

tren

d (re

duct

ion

by a

ppro

ximat

ely 5

%

ever

y yea

r). T

he m

ain ca

uses

of m

orta

lity f

rom

accid

enta

l poi

soni

ng in

Pol

and

are:

• 75

% –

accid

enta

l poi

soni

ng b

y alco

hol (

ICD

-10

code

X45

)

• 17

% –

accid

enta

l poi

soni

ng b

y oth

er g

ases

and

vapo

urs (

ICD

-10

code

X47

).

mor

talit

y fr

om a

ccid

enta

l poi

soni

ng b

y ag

e an

d ur

ban/

rura

l res

iden

cePo

land


KE

Y M

ESSA

GE:

The

risk

of d

eath

from

falls

is al

mos

t five

tim

es h

ighe

r for

male

s tha

n fe

male

s in

Rom

ania.

The

gro

ups a

t gre

ates

t risk

are c

hild

ren

aged

0–4

, male

s ove

r 45

year

s an

d fe

male

s ove

r 75

year

s.

Dat

a sou

rce:

Dat

a on

deat

hs b

y fa

lls ar

e rec

orde

d by

the R

oman

ian N

atio

nal I

nstit

ute o

f Sta

tistic

s, ba

sed

on th

e da

ta re

porte

d by

all

healt

h un

its a

nd p

rovid

ers.

Dat

a ar

e av

ailab

le at

the

natio

nal l

evel

at th

e Nat

iona

l Cen

tre fo

r Pub

lic H

ealth

Info

rmat

ion

and

Stat

istics

, with

in th

e Nat

iona

l Ins

titut

e of

Publ

ic H

ealth

. Dat

a stra

tified

by s

ex an

d ag

e gro

ups a

re av

ailab

le on

requ

est.

Dat

a de

scrip

tion:

Dea

ths c

ause

d by

falls

(IC

D-1

0 co

des W

10–W

19) a

re re

porte

d by

all

med

ical

prov

ider

s via

deat

h ce

rtific

ates

to th

e di

strict

offi

ces o

f the

Nat

iona

l Ins

titut

e of

Sta

tistic

s. D

ata

are

strat

ified

by s

ex an

d 5-

year

age g

roup

s in

cate

gorie

s fro

m 0

–4 to

85+

.

Dat

a cov

erag

e: T

he av

ailab

le da

ta h

ave n

atio

nal c

over

age f

rom

197

0. T

he IC

D-1

0 cla

ssifi

catio

n ha

s be

en u

sed

since

199

4. A

valid

atio

n pr

oces

s for

the m

ain ca

use o

f dea

th is

mad

e at t

he d

istric

t pub

lic

healt

h di

rect

orat

es.

Soci

al d

imen

sion:

Thi

s ine

quali

ty fa

ct sh

eet s

tratif

ies m

orta

lity

due

to fa

lls b

y se

x an

d ag

e gr

oups

, aim

ing

to a

sses

s the

mag

nitu

de o

f ine

quali

ty a

nd to

iden

tify

pote

ntial

targ

et g

roup

s for

pre

vent

ive

inte

rven

tions

.

Con

text

: A co

untry

asse

ssm

ent c

arrie

d ou

t by

the W

HO

Reg

iona

l Offi

ce fo

r Eur

ope o

n th

e sta

tus

of th

e im

plem

enta

tion

of R

egio

nal C

omm

ittee

reso

lutio

n EU

R/RC

55/R

9 an

d Eu

rope

an C

ounc

il re

com

men

datio

n 20

07/C

164

/01

on th

e pre

vent

ion

of in

jury

and

prom

otio

n of

safe

ty re

veale

d th

at

desp

ite th

e abs

ence

of a

nat

iona

l pol

icy fo

r inj

ury p

reve

ntio

n, se

vera

l pre

vent

ive m

easu

res a

re in

plac

e to

redu

ce n

umbe

rs of

falls

and

their

hea

lth c

onse

quen

ces.

The

se in

clude

regu

lator

y m

easu

res

for

child

-spe

cific

envir

onm

ents,

impr

ovem

ent o

f em

erge

ncy

resp

onse

s for

all

type

s of e

mer

genc

y, an

d he

alth

prom

otio

n in

itiat

ives o

n lif

e ski

lls, in

cludi

ng fa

ll pr

even

tion.

Nev

erth

eless

, the

resu

lts sh

ow th

at th

e nat

iona

l sta

ndar

dize

d m

orta

lity r

ate f

rom

falls

is h

ighe

r tha

n in

the E

U an

d th

e WH

O E

urop

ean

Regi

on, a

ccou

ntin

g fo

r abo

ut 1

400

prev

enta

ble d

eath

s per

year.

T

he re

cent

mor

talit

y tre

nd ev

en sh

ows a

n in

crea

se fr

om 6

.4/1

00 0

00 (2

005)

to 6

.8/1

00 0

00 (2

009)

, in

dica

ting

that

futu

re ac

tion

is ne

cess

ary.

Polic

y im

plic

atio

ns: A

ccor

ding

to th

e mor

talit

y dat

a, th

e mos

t vul

nera

ble g

roup

s are

child

ren

(esp

ecial

ly bo

ys) a

ged

0–4

year

s and

indi

vidua

ls ab

ove 5

5 ye

ars,

for w

hom

the l

evel

of ri

sk

stron

gly i

ncre

ases

with

risin

g ag

e, es

pecia

lly fo

r male

s. Sp

ecifi

c pub

lic h

ealth

inte

rven

tions

for t

hese

targ

et g

roup

s sho

uld

be d

esig

ned

and

impl

emen

ted.

Educ

atio

n an

d in

form

atio

n on

ris

ks, id

entif

icatio

n of

peo

ple a

t risk

, sec

onda

ry p

reve

ntio

n to

impr

ove a

dapt

abili

ty to

new

phy

sical

cond

ition

s, an

d sc

reen

ing

for h

igh-

risk

cond

ition

s and

visib

ility

or h

earin

g im

pairm

ent

shou

ld b

e un

derta

ken,

with

pro

mpt

seco

ndar

y in

terv

entio

n ac

cess

ibili

ty a

nd re

habi

litat

ion

proc

edur

es to

avo

id c

ompl

icatio

ns. T

o re

duce

the

mag

nitu

de o

f risk

of d

eath

by

falls

for

vuln

erab

le gr

oups

, sys

tem

atic

educ

atio

n of

thos

e re

spon

sible

for c

hild

ren

and

the

elder

ly sh

ould

also

be

impl

emen

ted.

Spec

ific

surv

eys a

nd b

ette

r mon

itorin

g of

the

circu

msta

nces

of

letha

l acc

iden

ts, to

geth

er w

ith a

syste

mat

ic va

lidat

ion

proc

edur

e for

codi

ng th

e cau

se o

f dea

th o

n th

e dea

th ce

rtific

ate,

coul

d co

ntrib

ute t

o th

e pro

visio

n of

fulle

r evid

ence

for i

nter

vent

ion.

Mor

talit

y fr

om fa

lls b

y ag

e an

d se

x (2

005

and

2009

)

0102030405060

0–4

5–1

4 1

5–24

25–

34 3

5–44

4

5–54

5

5–64

6

5–74

7

5–84

8

5+To

tal

Age

grou

p


Mal

es, 2

005

Fem

ales

, 200

5M

ales

, 200

9Fe

mal

es, 2

009

Ana

lysis

and

inte

rpre

tatio

n: T

he o

vera

ll cr

ude

deat

h ra

te fr

om fa

lls in

crea

sed

from

6.4

/100

000

in

200

5 to

6.8

/100

000

in 2

009,

signa

lling

a h

igh

and

incr

easin

g nu

mbe

r (ab

out 1

400

peop

le) o

f pr

even

tabl

e dea

ths.

The

risk

of d

eath

from

falls

is ab

out f

ive ti

mes

hig

her f

or m

ales t

han

for f

emale

s. T

he in

ciden

ce o

f fat

al fa

lls in

fem

ales s

light

ly de

cline

d fro

m 2

.7/1

00 0

00 in

200

5 to

2.5

/100

000

in

2009

.

The

age

dist

ribut

ion

of m

orta

lity

for 2

005

and

2009

reve

als in

crea

sing

valu

es fo

r male

s (to

tal m

ale

incid

ence

rose

from

10.

3/10

0 00

0 to

11.

3/10

0 00

0), e

spec

ially

in th

e cas

e of m

ales a

ged

55 ye

ars a

nd

abov

e. H

owev

er, th

ere i

s a co

nsta

nt p

rese

nce o

f thr

ee p

eaks

of r

isk th

at co

uld

be ta

ckled

:

• th

e 0–4

year

age g

roup

(bot

h se

xes)

• m

ales o

ver 4

5 ye

ars,

incr

easin

g wi

th ag

e

• fe

male

s ove

r 75

year

s, in

crea

sing

with

age.

mor

talit

y fr

om fa

lls b

y ag

e an

d se

xr

oman

ia


KE

Y M

ESSA

GE:

Ine

quali

ties

in t

raffi

c no

ise e

xpos

ure

by in

com

e lev

el ar

e sm

all b

ut

pres

ent i

n th

e Net

herla

nds.

Dat

a sou

rce:

Dat

a on

road

, rail

and

air

traffi

c no

ise e

xpos

ure

were

mod

elled

with

the

Dut

ch G

IS

noise

calcu

latio

n m

odel

EMPA

RA v

ersio

n 20

04-1

. In

addi

tion,

the

com

bine

d no

ise e

xpos

ure

from

th

ese

sour

ces

was

calcu

lated

. Inp

ut d

ata

are

avail

able

from

diff

eren

t Dut

ch d

ata

sour

ces

and

are

irreg

ular

ly up

date

d. O

utpu

t dat

a on

noi

se w

ere

com

bine

d wi

th d

ata

on h

ousin

g lo

catio

ns fr

om th

e A

ddre

ss C

oord

inat

es o

f the

Net

herla

nds d

atab

ase.

Dat

a des

crip

tion:

The

par

amet

er is

def

ined

as m

odell

ed tr

affic

noi

se le

vel a

t hom

e.

Dat

a cov

erag

e: D

ata a

re av

ailab

le fo

r the

who

le co

untry

. Inp

ut d

ata w

ere f

rom

200

0–20

02. F

or ro

ad

and

rail

traffi

c the

reso

lutio

n wa

s 25

× 25

m; f

or ai

r tra

ffic n

oise

the r

esol

utio

n wa

s 250

× 2

50 m

. The

no

ise le

vels

with

in th

ese g

rids w

ere a

ssig

ned

to al

l dwe

lling

s with

in th

at g

rid.

Soci

al d

imen

sion:

For

this

ineq

ualit

y fa

ct sh

eet n

oise

dat

a wer

e com

bine

d wi

th in

com

e dat

a at t

he

6-di

git p

ostc

ode l

evel

(abo

ut th

e siz

e of a

stre

et). I

ncom

e dat

a wer

e der

ived f

rom

the “

Geo

mar

ktpr

ofiel

” te

lepho

ne su

rvey

of D

utch

resid

ents

in 20

01, in

whi

ch pe

ople

were

aske

d to c

hara

cter

ize t

he do

min

ant

inco

me l

evel

in th

eir p

ostc

ode a

rea.

Con

text

: In

the d

ense

ly po

pulat

ed co

untry

of t

he N

ethe

rland

s a la

rge p

ropo

rtion

of t

he p

opul

atio

n (8

0%) i

s exp

osed

to tr

affic

noi

se o

f mor

e tha

n 50

dB(

A).

Seve

ral n

oise

mea

sure

s are

take

n to

redu

ce

noise

exp

osur

e, su

ch a

s no

ise b

arrie

rs, q

uiet

er r

oad

surfa

ces,

and

insu

latio

n of

hom

es. H

owev

er,

incr

easin

g tra

ffic n

oise

rem

ains a

pro

blem

, with

adve

rse ef

fect

s on

healt

h an

d we

ll-be

ing.

Dut

ch e

nviro

nmen

tal p

olicy

focu

ses o

n th

e ge

nera

l pop

ulat

ion

and

not o

n sp

ecifi

c su

bpop

ulat

ions

. Sp

atial

plan

s are

teste

d fo

r noi

se st

anda

rds.

The

se st

anda

rds a

re se

t by

the s

tate

to en

sure

a sa

fe an

d he

althy

livin

g en

viron

men

t for

all. T

he n

oise

dist

ribut

ion

amon

g th

e pop

ulat

ion

in th

e Net

herla

nds

is th

e res

ult o

f a co

mpl

ex in

tera

ctio

n of

pub

lic p

olicy

dec

ision

s, m

arke

t for

ces,

histo

rical

cont

ext a

nd

publ

ic pe

rcep

tions

. To

unde

rstan

d ex

posu

re d

iffer

ence

s am

ong

socio

econ

omic

grou

ps a

nd ta

ke th

e m

ost e

ffect

ive p

olicy

mea

sure

s it i

s im

porta

nt to

unr

avel

the c

ompl

ex u

nder

lyin

g m

echa

nism

s.

Polic

y im

plic

atio

ns: D

utch

envir

onm

enta

l pol

icy in

relat

ion

to sp

atial

plan

ning

ofte

n fo

cuse

s on

the g

ener

al po

pulat

ion

rath

er th

an sp

ecifi

c tar

get g

roup

s. T

he re

sults

abov

e sho

w th

at

such

an ap

proa

ch ca

n pr

ovid

e fair

ly eq

ual o

utco

mes

to th

e pop

ulat

ion,

as th

e diff

eren

ces b

etwe

en in

com

e cat

egor

ies ar

e sm

all. H

owev

er, in

equa

lities

may

grow

due

to th

e cur

rent

econ

omic

situa

tion

and

polit

ical c

limat

e. Lo

wer-

inco

me

cate

gorie

s ofte

n ha

ve fe

wer c

hoice

s abo

ut w

here

to li

ve a

nd a

re m

ore

frequ

ently

in w

eake

r hea

lth. I

nclu

ding

socia

l dim

ensio

ns in

spat

ial

plan

ning

wou

ld g

ive in

sight

into

the

effe

cts f

or d

iffer

ent i

ncom

e ca

tego

ries a

nd, i

f nec

essa

ry, e

nabl

e pl

anne

rs to

take

pre

caut

ions

. Fur

ther

mor

e, po

licy-

mak

ers s

houl

d ta

ke in

to a

ccou

nt

any

accu

mul

atio

n of

issu

es in

the n

eighb

ourh

ood

and

at w

ork

and

focu

s not

onl

y on

pro

blem

s, bu

t also

on

the a

men

ities

of t

he lo

cal e

nviro

nmen

t, su

ch as

qui

et an

d gr

een

spac

e. T

hese

en

viron

men

tal a

men

ities

may

impr

ove h

ealth

; for

exam

ple,

thro

ugh

stres

s red

uctio

n.

Traf

fic n

oise

exp

osur

e ab

ove

65dB

(A) i

n re

side

ntia

l are

as b

y ho

useh

old

inco

me

01234567

Roa

d tra

ffic

Rai

l tra

ffic

Air t

raffi

c C

ombi

ned

traffi

c


Hig

h in

com

eIn

com

e ab

ove

aver

age

Aver

age

inco

me

Low

inco

me

Min

imum

inco

me

Ana

lysis

and

inte

rpre

tatio

n: T

he d

ata

show

the

perc

enta

ge o

f hom

es p

er in

com

e ca

tego

ry w

ith a

m

odell

ed n

oise

leve

l hig

her t

han

65 d

B(A

) for

diff

eren

t sou

rces

of t

raffi

c no

ise (r

oad,

rail

and

air

traffi

c, an

d a c

ombi

natio

n of

thes

e). T

he re

sults

are m

ixed.

In ge

nera

l, the

diff

eren

ces b

etwe

en in

com

e ca

tego

ries a

re sm

all. L

ower

-inco

me c

ateg

ories

are e

xpos

ed to

slig

htly

high

er n

oise

leve

ls th

an h

ighe

r-in

com

e one

s, de

pend

ing

on th

e sou

rce o

f noi

se. D

iffer

ence

s are

mos

t pro

noun

ced

for r

ail tr

affic

; thi

s m

ay b

e bec

ause

wor

kers’

dwe

lling

s ten

ded

histo

ricall

y to

be b

uilt

close

to ra

ilway

s. Ba

sed

on re

gion

al an

alyse

s cov

erin

g ar

eas w

ith a

large

airp

ort (

not s

hown

her

e), h

ighe

r-in

com

e cat

egor

ies ar

e exp

osed

to

hig

her a

ir tra

ffic n

oise

leve

ls in

the

prox

imity

of l

arge

airp

orts.

Thi

s may

be e

xplai

ned

by th

e fa

ct

that

flig

ht p

aths

are p

lanne

d ov

er le

ss d

ense

ly po

pulat

ed ar

eas,

wher

e mor

e hig

h-in

com

e hou

seho

lds

live.

No

clear

tren

d wa

s fo

und

for

road

traf

fic. N

o ro

bust

expl

anat

ion

is av

ailab

le, b

ut it

may

be

assu

med

that

som

e hig

her-

inco

me h

ouse

hold

s live

clos

e to

busy

road

s for

acce

ssib

ility

reas

ons.

It m

ust b

e not

ed th

at m

odell

ed n

oise

expo

sure

bey

ond

65 d

B(A

) doe

s not

corre

spon

d en

tirely

to th

e pe

rcen

tage

of t

he p

opul

atio

n co

mpl

ainin

g ab

out n

oise

.

Expo

sure

to

traf

fic n

oise

wit

hin

resi

dent

ial a

reas

by

hous

ehol

d in

com

e le

vel

Net

herl

ands


KE

Y M

ESSA

GE:

Acc

ess t

o gr

een

spac

e va

ries s

igni

fican

tly a

mon

g di

ffere

nt re

gion

s in

Spain

. Reg

iona

l var

iatio

n str

ongl

y ex

ceed

s var

iatio

n by

inco

me

or so

cial c

lass.

A n

orth

–so

uth

distr

ibut

ion

is ap

pare

nt.

Dat

a sou

rce:

Dat

a on

the

lack

of a

cces

s to

gree

n sp

ace

are

take

n fro

m th

e N

atio

nal H

ealth

Sur

vey

carri

ed ou

t eve

ry fe

w ye

ars b

y the

Nat

iona

l Ins

titut

e of S

tatis

tics i

n co

llabo

ratio

n wi

th th

e Min

istry

of

Hea

lth, S

ocial

Pol

icy an

d Eq

ualit

y. The

repo

rted

data

for t

his f

act s

heet

are t

aken

from

the l

ast s

urve

y in

200

6 wh

ich in

clude

d 31

300

hom

es, d

istrib

uted

as 2

236

cens

us se

ctio

ns. M

etho

dolo

gy a

nd d

ata

can

be d

ownl

oade

d fro

m: h

ttp://

www.

msp

s.es/e

stadE

studi

os/e

stadi

stica

s/enc

uesta

Nac

iona

l/hom

e.ht

m.

Dat

a des

crip

tion:

The

par

amet

er is

def

ined

as “h

omes

with

lack

of a

cces

s to

gree

n sp

ace”

and

refe

rs to

the r

espo

nse g

iven

by m

embe

rs of

a ho

useh

old

to th

e sur

vey q

uesti

on “d

oes y

our h

ome h

ave o

ne o

f th

e fol

lowi

ng p

robl

ems?

” whe

re th

e pro

blem

liste

d wa

s “lac

k of a

cces

s to g

reen

spac

e”. T

he p

erce

ntag

e of

resp

onde

nts a

nswe

ring “

signi

fican

t pro

blem

”, “sl

ight

pro

blem

” or “

not a

pro

blem

” was

reco

rded

.

Dat

a co

vera

ge: T

he su

rvey

has

nat

iona

l cov

erag

e bu

t can

be

brok

en d

own

into

var

ious

terri

toria

l un

its su

ch as

the a

uton

omou

s com

mun

ities

, whi

ch re

pres

ent S

pani

sh re

gion

s.

Soci

al d

imen

sion:

Thi

s ine

quali

ty fa

ct sh

eet s

tratif

ies th

e pe

rcen

tage

of p

eopl

e re

porti

ng a

lack

of

acce

ss to

gre

en sp

ace p

er re

gion

in S

pain

. No

evid

ence

for a

socia

l dim

ensio

n of

this

ineq

ualit

y (su

ch

as b

y inc

ome)

was

foun

d.

Con

text

: Acc

ess t

o gr

een

spac

e ha

s bee

n fo

und

to b

enef

it m

any

aspe

cts o

f hea

lth a

nd w

ell-b

eing,

enab

ling

resid

ents

– pa

rticu

larly

thos

e in

urb

an a

reas

– to

cop

e be

tter w

ith th

e str

esse

s of l

iving

in

large

con

urba

tions

. Gen

erall

y, th

e pr

opor

tion

of g

reen

spa

ce p

er p

erso

n di

min

ishes

as

popu

latio

n de

nsity

incr

ease

s; th

us co

mpa

ct ci

ties s

how

very

low

per c

apita

gre

en sp

ace a

lloca

tion.

In S

pain

ther

e is

signi

fican

t var

iabili

ty in

the d

egre

e of u

rban

izat

ion

amon

g th

e reg

ions

.

A n

umbe

r of s

ocial

varia

bles

such

as in

com

e and

educ

atio

n lev

el we

re an

alyse

d bu

t did

not

expl

ain th

e va

riabi

lity

in a

cces

s to

gree

n sp

ace

amon

g au

tono

mou

s com

mun

ities

. Ana

lyse

d at

the

geog

raph

ical

level,

it w

as n

ot p

ossib

le to

expl

ain in

equa

lities

in ac

cess

to gr

een

spac

e in

term

s of a

socia

l dim

ensio

n.

How

ever,

ther

e se

ems t

o be

a c

lear n

orth

–sou

th sp

atial

pat

tern

, with

the

lowe

st pr

ovisi

on o

f gre

en

spac

e in

the r

egio

ns in

the s

outh

.

Polic

y im

plic

atio

ns: T

here

is a

signi

fican

t diff

eren

ce in

acce

ss to

gre

en sp

ace a

cros

s diff

eren

t reg

ions

in S

pain

. The

fact

ors d

eter

min

ing

this

varia

bilit

y can

not b

e elu

cidat

ed th

roug

h th

is an

alysis

, as t

he re

gion

s inc

lude

bot

h ur

ban

and

rura

l are

as, a

s well

as p

opul

atio

ns o

f var

ying

inco

me a

nd so

cial l

evels

. A h

ighe

r geo

grap

hica

l seg

rega

tion

and

inde

pend

ent s

tudy

of u

rban

an

d ru

ral a

reas

wou

ld en

able

a mor

e rig

orou

s ana

lysis

of t

he re

latio

n be

twee

n ac

cess

to g

reen

spac

e and

leve

l of i

ncom

e and

socia

l clas

s. A

t thi

s geo

grap

hica

l lev

el, th

ere w

ere n

o sig

nific

ant d

iffer

ence

s in

acce

ss to

gree

n sp

ace a

ccor

ding

to le

vel o

f inc

ome o

r soc

ial cl

ass a

s det

erm

ined

by o

ccup

atio

n. T

he ap

pare

nt n

orth

–sou

th

distr

ibut

ion

in ac

cess

to g

reen

spac

e may

thus

be i

nflu

ence

d by

per

capi

ta in

com

e, de

gree

of u

rban

izat

ion

or si

mpl

y clim

ate.

How

ever,

it in

dica

tes t

hat s

outh

ern

regi

ons m

ay n

eed

to b

e m

ore a

ctive

in p

rovid

ing

high

-qua

lity g

reen

spac

e to

the p

opul

atio

n.

Lack

of a

cces

s to

gre

en s

pace

in re

gion

s of

Spa

in

020406080100

Andalucía

Ceuta y Melilla

Canarias Communitat

Valenciana Murcia

Baleares

Extremadura Castilla-

La Mancha

Cataluña

Madrid PaísVasco

Galicia

Aragón

Rioja

Navarra Castilla

y León Asturias

Cantabria

Percentage of inhabitants

Not

a p

robl

em

Slig

ht p

robl

em

Sign

ifica

nt p

robl

em

S

outh

ern

regi

ons

N

orth

ern

regi

ons

Ana

lysis

and

inte

rpre

tatio

n: T

he p

robl

em o

f lac

k of

acc

ess t

o gr

een

spac

e am

ong

hous

ehol

ds in

di

ffere

nt re

gion

s pre

sent

s sig

nific

ant v

ariab

ility.

The

gra

ph p

rese

nts s

outh

ern

and

north

ern

regi

ons

(repr

esen

ting

the

auto

nom

ous

com

mun

ities

) in

dec

reas

ing

orde

r ac

cord

ing

to t

he c

ombi

ned

perc

enta

ge o

f res

pond

ents

repo

rting

“sig

nific

ant p

robl

em” o

r “sli

ght p

robl

em” o

f acc

ess t

o gr

een

spac

e an

d th

us in

ord

er o

f dec

reas

ing

prob

lem o

f lac

k of

acc

ess

to g

reen

spa

ce. I

t app

ears,

with

a fe

w ex

cept

ions

, tha

t the

per

cent

age

of re

spon

dent

s con

sider

ing

acce

ss to

gre

en sp

ace

to b

e a

prob

lem is

hi

gher

in (m

ostly

) sou

ther

n re

gion

s (su

ch a

s And

alucía

, Can

arias

, Com

unita

t Vale

ncian

a, M

urcia

, Ba

leare

s, Ex

trem

adur

a) t

han

in n

orth

ern

regi

ons

(such

as

Can

tabr

ia, A

sturia

s, C

astil

la y

Leon

, N

avar

ra, L

a Rio

ja, P

aís V

asco

). T

his c

ould

pos

sibly

be ca

used

by c

limat

ic di

ffere

nces

amon

g re

gion

s, wh

ere

mos

t so

uthe

rn r

egio

ns h

ave

warm

er a

nd d

rier

clim

ates

, hen

ce h

avin

g fe

wer

gree

n ar

eas.

How

ever,

this

woul

d no

t nec

essa

rily i

mpl

y a la

ck o

f acc

ess t

o na

tura

l spa

ce.

The

diff

eren

ces a

mon

g th

e re

gion

s cou

ld n

ot b

e ex

plain

ed in

term

s of a

socia

l dim

ensio

n. S

ever

al so

cial v

ariab

les su

ch a

s inc

ome

level,

edu

catio

n lev

el an

d so

cial c

lass a

s def

ined

by

occu

patio

n we

re

analy

sed,

but d

id n

ot ex

plain

regi

onal

diffe

renc

es. I

t app

ears

that

the r

egio

nal n

orth

–sou

th d

ispar

ity

is gr

eate

r th

an th

e di

spar

ity d

ue to

inco

me,

educ

atio

n or

soc

ial c

lass

of th

e re

spon

dent

s. H

ighe

r di

sagg

rega

tion

of d

ata b

y pro

vince

s wou

ld en

able

a mor

e det

ailed

analy

sis.

lack

of a

cces

s to

rec

reat

iona

l/gr

een

spac

e by

reg

ion

Spai

n


KE

Y M

ESSA

GE:

In

Ger

man

y, ch

ildre

n liv

ing

in h

ouse

hold

s wi

th lo

w SE

S ar

e m

ore

affe

cted

by t

obac

co sm

oke e

xpos

ure a

t hom

e tha

n th

ose w

ith h

ighe

r SES

.

Dat

a so

urce

: Dat

a on

the

pres

ence

of s

mok

ers

in th

e ho

useh

olds

of 1

790

child

ren

aged

3 to

14

year

s ar

e ta

ken

from

the

Fed

eral

Envir

onm

ent

Age

ncy’s

Ger

man

Env

ironm

enta

l Sur

vey

2003

/6

for C

hild

ren

(Ger

ES IV

); da

ta o

n SE

S fro

m th

e co

ncur

rent

Ger

man

Hea

lth S

urve

y fo

r Chi

ldre

n an

d A

doles

cent

s (K

iGG

S) s

tudy

by

the

Robe

rt K

och

Insti

tute

. Pub

lic u

se f

iles

with

dat

a fro

m

Ger

ES IV

and K

iGG

S are

avail

able

on re

ques

t via

http

://ww

w.ub

a.de/

surv

ey an

d http

://ww

w.ki

ggs.d

e, re

spec

tively

(in

Ger

man

).D

ata d

escr

iptio

n: In

que

stion

72

of th

e que

stion

naire

, eve

ry h

ouse

hold

mem

ber w

as li

sted

with

the

poss

ible

valu

es “s

mok

er” o

r “no

n-sm

oker

”. T

he to

tal n

umbe

r of s

mok

ers l

iving

in th

e hou

seho

ld w

as

sum

mar

ized

.D

ata c

over

age:

The

ava

ilabl

e da

ta o

f Ger

ES IV

are

repr

esen

tativ

e of

chi

ldre

n in

Ger

man

y. Fu

ture

su

rvey

s are

envis

aged

.So

cial

dim

ensio

n: T

his

ineq

ualit

y fa

ct s

heet

stra

tifies

the

pres

ence

of s

mok

ers

in th

e ho

useh

old

by S

ES. T

he S

ES in

dex

com

bine

s inf

orm

atio

n on

the

pare

nts’

educ

atio

nal a

nd p

rofe

ssio

nal s

tatu

s an

d on

the

hous

ehol

d’s n

et in

com

e. T

he lo

w SE

S gr

oup

com

prise

s app

roxim

ately

the

lowe

st 25

%,

the

mid

dle

SES

grou

p ap

prox

imat

ely th

e m

iddl

e 50

% a

nd th

e hi

gh S

ES g

roup

app

roxim

ately

the

high

est 2

5% o

f the

child

ren.

Con

text

: It i

s well

kno

wn th

at ch

ildre

n’s h

ealth

is af

fect

ed b

y par

enta

l sm

okin

g. C

hild

ren

of p

aren

ts wh

o sm

oke

show

, for

exa

mpl

e, an

incr

ease

d fre

quen

cy o

f sud

den

infa

nt d

eath

synd

rom

e, ac

ute

and

chro

nic i

nflam

mat

ion

of th

e mid

dle e

ar, as

thm

a and

infe

ctio

ns o

f the

lowe

r res

pira

tory

trac

ts. A

lmos

t on

e-ha

lf of

child

ren

aged

up

to 1

7 ye

ars a

re li

ving

in a

hou

seho

ld w

ith a

t lea

st on

e sm

okin

g pa

rent

(d

ata f

rom

KiG

GS

study

). W

ith th

e aim

of i

mpr

ovin

g th

e he

alth

outc

omes

for t

hese

chi

ldre

n, it

is im

porta

nt to

kno

w wh

ich

part

of t

he p

opul

atio

n is

mos

t af

fect

ed b

y (se

cond

-han

d) s

mok

e. W

ith t

his

know

ledge

, tar

gete

d in

form

atio

n an

d pr

even

tion

mea

sure

s ca

n be

take

n in

ord

er to

impr

ove

furth

er th

e pr

otec

tion

of

non-

smok

ers –

inclu

ding

child

ren.

En

cour

agin

g th

eir p

aren

ts to

ban

smok

ing

at h

ome

is lik

ely to

impr

ove

healt

h ou

tcom

es fo

r the

se

child

ren.

Polic

y im

plic

atio

ns: T

he b

est p

reve

ntio

n of

seco

nd-h

and

smok

e is a

low

rate

of t

obac

co co

nsum

ptio

n in

the w

hole

socie

ty. T

oget

her w

ith a

high

tax

on to

bacc

o, an

adve

rtisin

g ba

n on

to

bacc

o pr

oduc

ts an

d m

easu

res t

o pr

even

t peo

ple f

rom

star

ting

to sm

oke,

as w

ell as

mea

sure

s to

help

smok

ers t

o sto

p th

eir to

bacc

o co

nsum

ptio

n, sm

okin

g ba

ns ar

e of h

igh

impo

rtanc

e. To

bacc

o co

ntro

l mea

sure

s for

the w

orki

ng en

viron

men

t are

in p

lace,

and

since

Sep

tem

ber 2

007

smok

ing

has b

een

bann

ed in

all b

uild

ings

of t

he fe

dera

l gov

ernm

ent,

on p

ublic

tran

spor

t an

d at

pas

seng

er st

atio

ns.

In a

dditi

on, s

mok

ing

in re

staur

ants

and

othe

r pub

lic p

laces

has

bee

n ba

nned

(unf

ortu

nate

ly wi

th n

umer

ous e

xcep

tions

) in

all fe

dera

l sta

tes o

f Ger

man

y sin

ce th

e be

ginn

ing

of 2

008.

Har

mon

izat

ion

of th

e diff

eren

t law

s in

the f

eder

al sta

tes (

with

out e

xcep

tions

), as

well

as ta

rget

ed in

form

atio

n an

d pr

even

tion

mea

sure

s for

hig

hly

affe

cted

sect

ions

of t

he p

opul

atio

n –

such

as p

eopl

e/fa

mili

es w

ith a

low

SES

– wo

uld

help

to fu

rther

redu

ce to

bacc

o co

nsum

ptio

n in

socie

ty.

Smok

ers

in th

e ho

useh

old

of n

on-s

mok

ing

child

ren

by S

ES

38.7

52.4

66.0

39.0

29.7

24.0

22.418

.0

9.6

Low

SE

S

Mid

dle

SE

S

Hig

h S

ES

Per

cent

age

of h

ouse

hold

s

No

smok

erO

ne s

mok

erM

ore

than

one

sm

oker

Ana

lysis

and

inte

rpre

tatio

n: C

hild

ren

in l

ow S

ES f

amili

es a

re m

ore

pron

e to

be

expo

sed

to

envir

onm

enta

l tob

acco

smok

e. In

the G

erES

IV d

ata a

stro

ng so

cial g

radi

ent o

f num

ber o

f sm

oker

s in

the c

hild

ren’s

hou

seho

lds c

ould

be o

bser

ved:

acco

rdin

g to

the p

aren

t int

ervie

ws, 3

4% o

f the

hig

h SE

S no

n-sm

okin

g ch

ildre

n liv

e in

hous

ehol

ds w

ith o

ne o

r mor

e sm

oker

s, bu

t thi

s per

cent

age i

ncre

ases

to

61%

for t

he lo

w SE

S no

n-sm

okin

g ch

ildre

n.

The

fin

ding

s fro

m t

he q

uesti

onna

ire d

ata

were

con

firm

ed b

y ni

cotin

e an

d co

tinin

e (a

nico

tine

met

abol

ite) a

naly

sed

in u

rine,

as w

ell a

s be

nzen

e m

easu

red

in in

door

air.

The

ana

lyse

s re

veale

d a

disti

nct s

ocial

grad

ient f

or b

oth

cotin

ine i

n ch

ildre

n’s u

rine a

nd b

enze

ne in

the a

ir of

child

ren’s

room

s: th

e low

er th

e SES

, the

hig

her t

he p

ropo

rtion

of c

hild

ren

expo

sed

to to

bacc

o sm

oke (

cotin

ine l

evels

ab

ove t

he li

mit

of q

uant

ifica

tion:

70%

versu

s 32%

; mea

n of

ben

zene

: 2.5

6 µg

/m³ v

ersu

s 1.7

0 µg

/m³).

T

his r

eflec

ts th

e fac

t tha

t in

low

SES

fam

ilies

smok

ing

is m

ore p

reva

lent t

han

in fa

mili

es w

ith h

igh

SES.

Pote

ntia

l sm

oke

expo

sure

at

hom

e by

SES

Ger

man

y


KE

Y M

ESSA

GE:

The

200

5 law

ban

ning

smok

ing

in in

door

pub

lic p

laces

in It

aly a

lso

affe

cted

exp

osur

e to

sec

ond-

hand

tob

acco

sm

oke.

Nev

erth

eless

, diff

eren

ces

in s

econ

d-ha

nd sm

oke e

xpos

ure r

emain

, and

are e

spec

ially

stron

g in

wor

kplac

es.

Dat

a so

urce

: Dat

a ar

e ta

ken

from

the

Euro

baro

met

er su

rvey

on

toba

cco

(200

9), a

nd th

e na

tiona

l an

nual

surv

ey o

n sm

okin

g co

nduc

ted

by D

OXA

, the

Ita

lian

bran

ch o

f the

Gall

up I

nter

natio

nal

Ass

ociat

ion,

and

the I

talia

n N

atio

nal I

nstit

ute o

f Hea

lth.

Dat

a des

crip

tion:

Dat

a on

expo

sure

to to

bacc

o sm

oke a

t the

wor

kplac

e are

give

n as

a pe

rcen

tage

of

resp

onse

s to

the q

uesti

on “h

ow o

ften

are y

ou ex

pose

d to

toba

cco

smok

e ind

oors

at yo

ur w

orkp

lace?

” Su

rvey

resp

onde

nts a

re a

repr

esen

tativ

e sam

ple o

f the

tota

l pop

ulat

ion

over

14

year

s.

Dat

a cov

erag

e: D

ata

relat

e to

Eur

obar

omet

er d

ata

of 2

009.

Repo

rted

data

by

DO

XA a

re a

naly

sed

for t

he p

erio

d 20

05–2

009.

The

que

stion

naire

s hav

e nat

iona

l cov

erag

e.

Soci

al d

imen

sion:

Thi

s ine

quali

ty fa

ct sh

eet a

naly

ses e

xpos

ure t

o to

bacc

o sm

oke a

t the

wor

kplac

e by

sex,

age a

nd S

ES. T

he d

ata s

how

SES

strat

ifica

tion

by se

lf-po

sitio

ning

on

a soc

ial sc

ale, a

s well

as b

y oc

cupa

tiona

l pos

ition

, whi

ch re

flect

s a si

mila

r situ

atio

n.

Con

text

: A n

ew la

w im

posin

g a

ban

on sm

okin

g in

pub

lic a

nd a

t wor

kplac

es, a

llowi

ng it

onl

y in

sp

ecial

ly de

signa

ted

spac

es, c

ame i

nto

forc

e in

Italy

on 1

0 Ja

nuar

y 200

5. T

he la

w ha

s had

subs

tant

ial

imm

ediat

e ef

fect

s, lea

ding

to

appr

oxim

ately

90%

of

publ

ic pl

aces

bec

omin

g sm

oke-

free.

For

work

plac

es o

ther

than

hos

pita

lity

venu

es, t

his p

erce

ntag

e dec

reas

es fr

om 9

0 to

70

(acc

ordi

ng to

the

DO

XA su

rvey

s of 2

005,

2006

, 200

7, 20

08 an

d 20

09).

Whi

le th

e pe

rcen

tage

of s

mok

e-fre

e pl

aces

has

rem

ained

con

stant

ove

r tim

e fo

r wor

kplac

es (a

nd

sligh

tly d

ecre

ased

to 8

5% fo

r pub

lic p

laces

), th

e Ita

lian

popu

latio

n’s fa

vour

able

attit

ude

towa

rds t

he

smok

ing

ban

law h

as in

crea

sed

over

tim

e (Eu

roba

rom

eter

dat

a).

Alth

ough

the i

mpa

ct o

f a la

w ap

plyi

ng u

nive

rsally

to ev

ery w

ork

envir

onm

ent w

as ex

pect

ed to

be t

he

equa

lizat

ion

of e

xpos

ure

for a

ll so

cial c

ateg

ories

of w

orke

rs, re

porte

d da

ta in

dica

te th

at d

iffer

ence

s re

main

.

Polic

y im

plic

atio

ns: A

smok

ing b

an is

a go

od p

olicy

for c

ontro

lling

expo

sure

to se

cond

-han

d sm

oke a

t wor

k, an

d sh

ould

also

be e

ffect

ive in

tack

ling s

ocial

ineq

ualit

ies in

smok

e exp

osur

e at

wor

k, du

e to

the h

ighe

r pre

valen

ce o

f sm

oker

s am

ong

blue

-col

lar w

orke

rs.D

espi

te th

e suc

cess

ful I

talia

n ba

n on

smok

ing

in p

ublic

and

work

plac

es o

f 200

5, sig

nific

ant s

ocial

ineq

ualit

ies in

expo

sure

to sm

oke a

t wor

k we

re st

ill o

bser

ved

by 2

009:

amon

g 30

% o

f wo

rkpl

aces

did

not

com

ply w

ith th

e law

. Low

-inco

me,

man

ual a

nd se

lf-em

ploy

ed w

orke

rs ar

e mos

t affe

cted

by c

ontin

ued

expo

sure

.T

hese

disa

ppoi

ntin

g re

sults

in th

e wor

kplac

e sug

gest

the n

eed

to d

evelo

p se

lectiv

e pro

gram

mes

to en

sure

enfo

rcem

ent o

f the

law

in o

ccup

atio

nal e

nviro

nmen

ts.

Expo

sure

to s

econ

d-ha

nd to

bacc

o sm

oke

at w

ork

(200

9)

01020304050

Male

Female

15–24

25–39

40–54

55+

Low

Medium

High

Self-employed

Managers

Otherwhitecollar

Manualworkers

Sex

Age

SES

(sel

f-pos

ition

ing)

Occ

upat

iona

l po

sitio

n

Percentage of working population Ana

lysis

and

inte

rpre

tatio

n: E

urob

arom

eter

dat

a sh

ow la

rge

ineq

ualit

ies in

exp

osur

e to

sec

ond-

hand

smok

e at

wor

k ac

cord

ing

to b

oth

SES

and

occu

patio

nal s

cale,

in fa

vour

of m

ore

adva

ntag

ed

posit

ions

. The

dat

a re

fer t

o re

spon

dent

s citi

ng a

ny p

erio

d of

exp

osur

e (le

ss th

an 1

hou

r, 1–

5 ho

urs,

mor

e tha

n 5

hour

s).

The

pol

lutio

n lev

els fr

om se

cond

-han

d sm

oke i

n pu

blic

and

at w

orkp

laces

hav

e bee

n re

duce

d by

the

law, a

s dem

onstr

ated

by

loca

l stu

dies

on

parti

culat

e m

atte

r with

an

aero

dyna

mic

diam

eter

small

er

than

2.5

µm

(PM

2.5)

and

ultra

fine p

artic

les an

d ni

cotin

e con

cent

ratio

ns in

indo

or ai

r of p

ublic

plac

es,

and

on th

e pr

esen

ce o

f urin

ary

cotin

ine

in n

on-s

mok

ing

empl

oyee

s of t

hese

wor

kplac

es, b

ut th

eir

socia

l dist

ribut

ion

is un

know

n. L

ocal

studi

es o

n ho

spita

l adm

issio

ns fo

r hea

rt fa

ilure

sugg

est t

hat

smok

ing

bans

may

also

be

effe

ctive

in r

educ

ing

the

impa

ct o

f se

cond

-han

d sm

oke

expo

sure

on

coro

nary

hea

rt di

seas

e and

nar

rowi

ng it

s soc

ioec

onom

ic di

strib

utio

n.

How

ever,

socia

l ine

quali

ties i

n ex

posu

re to

seco

nd-h

and

smok

e at

wor

k ar

e sti

ll ob

serv

ed in

Itali

an

Euro

baro

met

er d

ata a

t the

end

of 2

009,

prob

ably

due t

o a s

elect

ive im

plem

enta

tion

of th

e ban

, with

lo

wer c

ompl

iance

amon

g se

lf-em

ploy

ed an

d m

anua

l wor

kers.

Expo

sure

to

seco

nd-h

and

smok

e at

wor

k by

sex

, age

, inc

ome,

and

em

ploy

men

tIt

aly

2

EXAMPLES OF NATIONAL PRACTICES IN ANALYSIS

AND PRESENTATION OF ENVIRONMENTAL

HEALTH INEQUALITIES

ANNEX

Annex 2152

ANNEx 2. ExAmPlES OF NATIONAl PrACTICES IN ANAlySIS ANd PrESENTATION OF ENvIrONmENTAl HEAlTH INEquAlITIES

CONTENTS

Environmental health inequality action in France: a report on the SIGFRIED Project�� 153

The United Kingdom sustainable development indicators: reporting on environmental inequalities �� 158

Identification of environmental health inequalities in populations living close to waste disposal sites in Italy �� 161

Assessing and reporting on inequalities related to lack of heated space and indoor pollution in Serbia and Montenegro �� 164

Analysis and presentation of environmental health inequalities concerning “green space” in Scotland �� 168

Assessment of environmental health inequalities in Finland: residential exposure to ambient air pollution from wood combustion and traffic �� 171

Environmental health inequality report on the impact of SES on the prevalence of allergies and respiratory diseases and symptoms in Hungarian children �� 175


ENvIrONmENTAl HEAlTH INEquAlITy ACTION IN FrANCE: A rEPOrT ON THE SIGFrIEd PrOjECT

INTrOduCTIONAnalysing the associations between the environment and health has become a major issue for public health in France, brought into focus by the country’s National Environmental Health Action Plans (NEHAPs). The second NEHAP of 2009 was prepared in response to the Fourth Ministerial Conference on Environment and Health, organized by the World Health Organization in 2004. Environmental inequality has become a fundamental theme that guides policy developments in France and the NEHAP has been adopted in every region.

The SIGFRIED Project, led by INERIS (the French National Institute for Industrial Environment and Risks), has been set up to identify and stratify environmental indicators and to map environmental disparities using spatial analysis techniques. The indicators bring together environmental and population data for several research applications:• to highlight vulnerable areas with significantly elevated exposure risk indicators in order to define

environmental monitoring campaigns, to manage and plan remedial actions;• to map environmental disparities throughout France;• to provide environmental indicators to quantify spatial relationships between environment, disease

and socioeconomic data.

This project uses the GIS-based modelling platform PLAINE (Environmental inequalities analysis platform), which allows the ongoing systematic collection, integration and analysis of data on emission sources, environmental contamination, exposure to environmental hazards, population and health.1

The Equit’area Project, led by EHESP (the School of Public Health), is investigating the association between infant mortality, deprivation and proximity to polluting industries.2 A specific deprivation index was calculated using the smallest administrative unit for which socioeconomic and demographic information is available from the National Institute of Statistics and Economic Studies (INSEE) 1999 census data.

dATA ANd mETHOdOlOGyThe case study combined information from national databases and a spatial approach to increase the effectiveness of the maps used for planning and decision-making on safeguarding public health. A spatial database was assembled from a set of variables characterizing environmental and population data. This approach integrated the emissions register and census data onto the map and analysed the spatial relationships between emissions sources and socioeconomic indicators. The following examples show the results produced by applying this method to data from the Nord-Pas-de-Calais region of France, using different administrative scales depending on objectives and data availability.

1 Caudeville J et al� (2011)� Construction d’une plate-forme intégrée pour la cartographie de l’exposition des populations aux substances chimiques de l’environnement� Environnement, Risques et Santé, 10(3):239–242�

2 Padilla C et al. (2011). Mortalité infantile, défaveur et proximité aux industries polluantes: une analyse spatiale conduites à fine échelle (agglomération de Lille, France)� Environnement, Risques et Santé, 10(3):216–221�

Annex 2154

Emission source data (road traffic and industrial) from national databases were used to build a proximity indicator to pollution sources as a proxy for exposure (see Fig. 1). Census data were used to build socioeconomic indicators (see Fig. 2). Proximity indicators were then stratified on the demographical information and spatial relationships between environmental and socioeconomic data were mapped and analysed (see Fig. 3).

road traffic and industrial site proximity indicatorsTwo main data sources were used.• The industrial facilities data used in the study were extracted from the French Register of Pollutant

Emissions from 2003 to 2009. The collection of these data meets the requirements of the EU protocol on Pollutant Release and Transfer Registers (PRTR), signed in Kiev in 2003, and those of the EU regulation concerning the establishment of a European PRTR (Regulation (EC) No. 166/2006, 18 January 2006).

• Emission source data on road locations and traffic capacity were compiled using a georeferenced national emissions database: the National Spatialized Inventory.

Emission source data were collected, harmonized and stored on a GIS. Distance-to-source indicators (proxy) were built using adapted distances and a GIS-based buffer tool (see Fig. 1). Buffer zones around sources were generated using different distances related to emission type, such as 1 km from industrial sites and 200 m from roads, and were used to create the proximity indicators.

Proxies were then aggregated using referent grid or administrative boundaries. Analysing the relationships between environmental and population indicators means that data could be presented under a common denominator, such as their spatial location or distribution. This was achieved by depicting the different data as layers and superimposing these layers in the same geographical system. In the above example, data were normalized by surface unit and aggregated on county boundaries using geometric ratios.

Socioeconomic indicatorsSeveral data sources were used:• administrative boundary information, varying according to the unit scale of analysis• population data from INSEE 2006 census data.

The Townsend deprivation index presented here (see Fig. 2) was calculated using four variables from data provided by routine census surveys:• economically active people unemployed• households with more than one person per room• households without a car• households not owner-occupied.

Negative values of the overall Townsend scores reflect less deprived areas; positive values reflect more deprived areas.

Socioeconomic indicators (also called deprivation indices) are widely used in public health, both in epidemiological analysis and in allocation of resources.


Fig. 1. Areas with proximity to industrial sites and principal roads using buffer zones

Source: INERIS.

Fig. 2. mapping the Townsend index on the case study area

Source: Nord-Pas-de-Calais Regional Health Observatory.

rESulTS

Stratification of environmental indicators on population dataEnvironmental indicators were aggregated with population data (in this case, stratified by age and sex) to quantify and map population under risk from exposure. Fig. 3 locates (a) and quantifies (b) residential populations close to industrial sites (within 1 km). In this example, around 16% of the population in the Nord-Pas-de-Calais region (700 000 inhabitants) is living in close proximity to an industrial site. A similar approach was used to aggregate the traffic map with the demographic data.

Annex 2156

Fig. 3. Aggregation of industrial indicators with gender and age subgroup data

Source: INERIS.

Other types of aggregation could also be studied, such as socioeconomic subgroups or income levels.

Analysing relationships between environmental and socioeconomic indicesEnvironmental and socioeconomic indices were then analysed using different spatial scales in order to assess inequality relationships. The bivariate local Moran test identifies spatial patterns in two variables. The black areas in Fig. 4(a) show correlations between high-risk environmental and socioeconomic indicators at the county level: in these areas, the strongest inequalities are detected for both deprivation and proximity to industrial sites. The white clusters show areas where the results are low for both indicators, and the grey areas show negative correlations or outliers.

Fig. 4. bivariate local moran results and Equit’area Project results

(a) Source: INERIS. (b) Source: EHESP.

A relevant spatial scale is essential to capture and identify inequality phenomena. Other approaches can also be used in order to refine this analysis. The Equit’area Project used the smallest administrative unit in France for which socioeconomic and demographic information is available from the national census. An index was constructed using a principal component analysis from a selection of 20 socioeconomic and demographic variables, reflecting multiple dimensions of socioeconomic deprivation such as income, education attainment, employment, housing characteristics, family structure, and immigration status.


The study setting was the black area highlighted by the local Moran test: the Lille metropolitan area, an urban area of 85 municipalities (about 612 km²). As shown in Fig. 4(b), most polluting industries identified are located in areas with a high deprivation score.

CONCluSIONSThe examples presented here show how GIS can be used to build indices in order to map and analyse environmental and socioeconomic inequalities. The initial screening approach enables identification of a vulnerable area where a more refined approach is developed at a smaller scale. In this particular aggregation, the results demonstrate that most of the polluting industries are localized in areas with a high deprivation score.

The geography of census data clearly has a fundamental influence on any study for which the data are used. Associations observed at the regional level applied to individuals within the regions can lead to a so-called ecological fallacy. Nevertheless, those approaches allow a quantification of relationships in order to identify areas that accumulate different types of inequalities, such as health, environmental and socioeconomic.

In the SIGFRIED Project, strong correlations were also found between cancer and socioeconomic indicators in this region. All these results facilitate the management and planning of targeted response actions in order to minimize health inequalities.

Annex 2158

THE uNITEd KINGdOm SuSTAINAblE dEvElOPmENT INdICATOrS: rEPOrTING ON ENvIrONmENTAl INEquAlITIES

INTrOduCTIONIn 2005 the United Kingdom committed, as part of its sustainable development strategy, to report on environmental inequalities within its measures of sustainable development.

A methodology was developed through a series of studies examining environmental quality indicators related to economic and social data linked to specific geographical areas nationwide. As part of this project, the Environment Agency with the Department of the Environment, academic researchers and nongovermental organizations developed a list of environmental quality variables to be imported into a GIS tool known as the Environmental Quality Index (see Table 1 for the variables covered). The GIS tool provides an illustration to different users, often working at different geographical scales, of the relative environmental quality of an area compared to national and local averages.

The geographical display of different environmental indicators alongside social and economic data for a given area provides users with an ability to articulate their own priority interests for reporting. An annual snapshot of environmental equality was also generated, illustrating as broadly as possible the spread of environmental quality across the social gradient.

dATA ANd mETHOdOlOGyIn order to develop a single indicator summarizing overall environmental inequality, the limitations of the data would need to be accepted in moving from a highly involved and flexible GIS to a single graph that captured a fair reflection of the data presented.

Given the spatial resolution of the data, broken down to areas with an average number of 1500 residents, a rationalization of what could be illustrated was required. Researchers and the government were also keen to avoid stigmatising areas based on nationally aggregated data, so the data was displayed against deciles (groups of 10%) of the population broken down by deprivation rather than spatially.

Deprivation indices are an approach to measuring social and economic disadvantage in the United Kingdom, derived from over 30 years of work by the government to provide policy-makers with information on the geographic nature of social and economic policy priorities. As such, across the different countries within the United Kingdom, locally determined indices of multiple deprivation have been developed and published periodically by governments in England, Wales, Scotland and Northern Ireland.3

Across these different geographical areas different methodological approaches have been selected that articulate the issues pertaining to the different countries. Evidently, the geography of these nations varies markedly, as does the geographical scale at which policy interventions are made.

3 More information regarding indices of multiple deprivation is available from the Office of National Statistics (www.ons.gov.uk) and from the Department for Communities and Local Government (www�imd�communities�gov�uk)�


Table 1. Index of multiple deprivation and environmental quality spatial statistics available in England

Index of multiple deprivation – living Environment

Indoor living Central heatingHousing condition

Outdoor living Road traffic accidentsCombined air quality indicators Nitrogen dioxide

Particulate matter (PM10)Sulphur dioxideBenzene

Environmental quality indicators

Proximity to regulated sitesCarbon dioxide emissions Ecological footprintDerelict landFly tippingLitter DetritusGraffitiGreen spaceBiodiversityFlood riskRiver water quality

As an example, an indicator of environmental equality has been developed for England as a geographical area that covers 11 environmental conditions or characteristics: river water quality, air quality, green space, biodiversity, flood risk, litter, housing condition, road accidents and the presence of regulated sites, including waste management sites, landfill sites or sewage treatment works. Fig. 5 below illustrates – for the deprivation deciles representing high to low levels of the English Indices of Multiple Deprivation – the degree to which environmental conditions affect these populations.

Fig. 5. Environmental conditions by deprivation deciles in England

0

10

20

30

40

50

60

70

80

90

100

Level of deprivation

Pre

vale

nce

(%)

4 or more conditions 3 conditions2 conditions1 condition

Mostdeprived

areas

Leastdeprivedareas

Source: Department for Environment, Food and Rural Affairs, Environment Agency (England & Wales), Department for Communities and Local Government.Note: for each of the environmental conditions, the population living in areas with, in relative terms, the 10% least favourable conditions have been determined.

Annex 2160

rESulTSAs is evident from Fig. 5 above, there is a clear bias toward improved environmental quality among populations that are also less socially and economically disadvantaged (or deprived). This general trend has emerged in each publication of the data, despite the adjusted list of environmental conditions measured (as a result of data availability and spatial coverage).

This has resulted in some significant findings for policy-makers.• It is clear that people in deprived areas experience greater exposure to air pollution, flooding, close

proximity to large industrial and waste management sites and poor river water quality. The evidence shows that the poorer the population, the more exposed it is to more unfavourable environmental conditions.

• Around 0.2% of people living in the least deprived areas may experience four or more environmental conditions that are “least favourable”. This rises to around 17% of people living in the most deprived areas.

CONCluSIONSIn June 2011, the United Kingdom Government produced a 60-year vision for the natural environment in England, publishing a White Paper entitled The natural choice: securing the value of nature.4 In this high-level policy document the indicator of environmental quality was presented to illustrate not only the presence of environmental inequality in England, but also to endorse the value of environmental improvement and good stewardship to promote better and healthier lives in England. The indicator of environmental quality has thus been exemplary in helping to set clear policy goals for the government and its partners to improve health and well-being and the environment.

4 http://www�defra�gov�uk/environment/natural/whitepaper/, accessed 21 September 2011�


IdENTIFICATION OF ENvIrONmENTAl HEAlTH INEquAlITIES IN POPulATIONS lIvING ClOSE TO wASTE dISPOSAl SITES IN ITAly

INTrOduCTIONThe patterns of associations between proximity to waste disposal sites and social deprivation, based on international data, have recently been reviewed.5 Similar assessments are being worked on in Italy, and several projects have been developed at both regional and local levels to assess the health effects of exposure to environmental pollution from waste disposal facilities, mainly incinerators and landfills. The Moniter Project in the Emilia–Romagna region (www.moniter.it), the ERAS Project (www.eraslazio.it) in the Lazio region, and various studies performed in the Campania region provide useful information on the measurement of inequality in waste site exposure.

At the national level, within the EU Integrated Assessment of Health Risks of Environmental Stressors in Europe Project (INTARESE), the “Waste” work package, coordinated by Dr Francesco Forastiere of the Lazio Region Health Service, has developed indicators for the health impact of pollution from the entire process of disposal of municipal solid waste, collecting data from three different countries (Italy, Slovakia and England) for a total of 905 municipal urban solid waste landfills and 53 waste incinerators. The following example refers to the work on Italy within this project.6 A project funded by the Italian Ministry of Health is currently considering the health-related issues of waste, with the aim of characterizing population exposures in different regions.

dATA ANd mETHOdOlOGyCountry-specific sources of information were identified and data about Italian municipal solid waste production and management were provided by the Italian Environmental Protection Agency (APAT). Data were checked for consistency and integrated with information collected from an independent body, the National Waste Observatory. Local databases were used where available to improve information on the waste facilities. The assessment referred to the situation in 2001.

APAT provided a database of information on incinerators operating during the period 2001–2007. In addition, a detailed census of the 52 incinerators operating in 2005 was made by the National Agency for New Technologies, Energy and Sustainable Economic Development, funded by the National Association of Stakeholders in Waste Management (Federambiente).

APAT also provided a database of landfill sites in Italy (a total of 619 in 2001) with information on the total capacity and waste land filled per year. Unfortunately, the database did not contain GIS coordinates and it was impossible to retrieve this information for the entire country. Using contacts with regional environmental authorities, geographical information could be obtained for five regions: Piemonte and Emilia–Romagna (in the north), Toscana and Abruzzi (in the central region), and Campania (in the south), covering a total of 118 landfills. For the rest of the country, it was assumed (with a moderate level of confidence) that the characteristics (sex, age and SES) of the population around the 501 missing landfills were similar to those of the 118 studied sites.

5 Martuzzi M, Mitis F and Forastiere F (2010)� Inequalities, inequities, environmental justice in waste management and health� European Journal of Public Health, 20(1):21–6�

6 Forastiere F et al� (2011)� Health impact assessment of waste management facilities in three European countries� Environmental Health, 10:53�

Annex 2162

Census tracts are the smallest units of aggregated population data that are available and geocoded in Italy. There are 352 138 census tracts, with 162 residents on average in each (ranging from 0 to 3386 people). GIS coordinates of all incinerators and landfills were collected where available (country-wide for incinerators and within five regions for landfills), and the population living within a specific radius of both incinerators and landfills was classified according to the census tract and its deprivation value. For each census tract in Italy, a deprivation index was available from a national project funded by the Ministry of Health,7 summarizing five dimensions of socioeconomic position (education, occupation, home ownership, family composition and nationality), by mean of an unweighted average of z-scores of each factor. For the purpose of this study, the corresponding population was distributed in quintiles of deprivation, ranging from very well off (level I) to very underprivileged (level V).

The distance from the point source (landfill site and/or incinerator) was used to estimate the exposed population. Buffer zones of 3 km around incinerators8 and 2 km around landfill sites9 were used as the likely limits of the dispersion of emissions. For each plant (incinerators and landfills), increasing distance (1, 2 and 3 km) from the centre (the formal address of the plant) was defined and the census tracts that matched these areas were included in the study.

rESulTSTable 2 shows the characteristics of the populations living close to the incinerators in Italy. More than a million people are affected, including 9010 neonates. More than 19% of the affected population is aged 65+ and the socioeconomic distribution is skewed towards higher deprivation (25% in class V (deprived) versus 12.6% in class I (less deprived)). More than half of the affected population lives close to plants built after 1990. The majority of residents within 3 km (64.4%) are located in the 2–3 km buffer zones.

Table 2. Characteristics of residents living within 3 km of an incinerator in Italy, 2001

variable N %

Total 1 060 569Sex M 511 831 48�3 F 548 738 51�7Age 0 9 010 0�8 1–14 123 061 11�6 15–44 435 825 41�1 45–64 289 430 27�3 65+ 203 243 19�2SES I 133 211 12�6 II 159 735 15�1 III 223 059 21�0 IV 257 009 24�2 V 264 401 24�9 Missing 23 154 2�2distance 0–1 km 50 990 4�8 1–2 km 326 798 30�8 2–3 km 682 781 64�4

7 Caranci N et al. (2010). The Italian deprivation index at census block level: definition, description and association with general mortality. Epidemiologia e Prevenzione, 34(4):167–76�

8 Elliott P et al� (1996)� Cancer incidence near municipal solid waste incinerators in Great Britain� British Journal of Cancer, 73(5):702–10�9 Elliott P et al. (2001). Risk of adverse birth outcomes in populations living near landfill sites. British Medical Journal, 323:363–8�


Table 3 shows the characteristics of the population living close to the 118 landfills in the five regions with geocoded data, and the corresponding figures for the whole country, assuming that the characteristics (sex, age and SES) of people around the 501 missing landfills were similar to those of the 118 studied sites. In Italy, more than 1 350 000 people would be affected, including 11 766 neonates. More than 18% of the population is aged 65+ and the socioeconomic distribution is skewed towards higher deprivation (26% in class V versus 13% in class I). The majority of residents within 2 km (85.7%) are located in the 1–2 km buffer zone.

Table 3. Characteristics of residents living within 2 km of a landfill site in Italy, 2001

variable N (observed data in five regions)

% N (estimated data at the national level)

Total 257 513 1 350 852

Sex M 125 750 48�8 659 655 F 131 763 51�2 691 197Age 0 2 243 0�9 11 766 1–14 32 801 12�7 172 066 15–44 107 244 41�6 562 577 45–64 67 971 26�4 356 560 65+ 47 254 18�4 247 883SES I 34 252 13�3 179 678 II 38 715 15�0 203 090 III 57 801 22�4 303 210 IV 59 320 23�0 311 179 V 67 339 26�1 353 244 Missing 86 0�0 451distance 0–1 km 36 716 14�3 192 603 1–2 km 220 797 85�7 1 158 249

CONCluSIONSA direct relationship between small area deprivation and residence near incinerators and landfills was found in Italy. Population estimates of social inequalities in exposure to residence near landfills are less reliable at the national level, due to the lack of complete information about the accurate location of plants.

These findings provide useful information to take into account, together with other factors such as the transport of waste from sources to plants, that could contribute significantly to the overall impact of these facilities.

Annex 2164

ASSESSING ANd rEPOrTING ON INEquAlITIES rElATEd TO lACK OF HEATEd SPACE ANd INdOOr POlluTION IN SErbIA ANd mONTENEGrO

INTrOduCTIONPoverty reflects a lack of household capital: households are poor if they are not able to meet their basic needs or secure a healthy lifestyle. Households’ strategies for coping with their inability to obtain energy services often have unwelcome consequences. Especially poor households practise a risky form of energy saving to reduce costs, such as using cheap and low-quality fuels, reducing the number of heated rooms and using fuels, such as unseasoned wood, that produce indoor pollution.

This work presents some evidence of the influence of poverty and social inequality on health in Serbia and Montenegro. The assessment was performed through the United Nations Development Programme (UNDP) during 2002 and 2003, investigating the energy sector in Serbia and Montenegro.10

dATA ANd mETHOdOlOGyThree surveys and a series of focus group interviews were conducted to understand the structure of energy services in Serbia and Montenegro. Two surveys examined households: the first examined actual energy usage patterns (consumption, appliances) and the second explored perceptions about energy processes and services. The perception survey was conducted to test the reliability of the findings of the energy usage survey and to provide additional insights into energy consumption patterns as well as people’s perceptions, needs, apprehensions and expectations about energy. A third survey looked at energy providers and auxiliary services. A series of focus group interviews was conducted with energy users to gain a deeper understanding of the complex processes examined in the surveys.

The energy usage survey was administered to a random sample of 1720 households. Municipalities, neighbourhoods, streets, houses and apartments were randomly selected. The perception survey was conducted on a separate sample of 1650 households randomly selected from the 2002 census in Serbia and voting lists in Montenegro. All findings were compared with the results of the Living Standards Measurement Study survey conducted as part of Serbia and Montenegro’s Poverty Reduction Strategy Process, which sampled more than 6500 households, and the household consumption survey conducted by the Federal Statistical Office. The results of the household survey were used for a cluster analysis to distinguish groups of households based on their energy patterns and welfare status.

The energy usage survey reveals that during the 2002/2003 heating season, the most frequently used fuels were wood and coal (61% of households), district heating (including central heating) (22%) and electricity (12%). In urban settlements, almost equal numbers of households used wood and coal (40%) and district heating (38%), while 16% of households used electricity as the primary fuel for heating. In rural settlements, 87% of households used wood and coal, and only 7% used electricity as the primary fuel for heating. Gas was used as a heating fuel mostly in Vojvodina (13% of households). In Belgrade almost half of households (47%) used district heating.

10 UNDP (2004)� Stuck in the past: energy, environment and poverty in Serbia and Montenegro� Belgrade, UNDP (http://www�undp�org/energy/docs/Stuck_in_the_Past�pdf, accessed 21 September 2011)�


INEquAlITIES rElATEd TO HEATING The surveys revealed that many households in Serbia and Montenegro lack sufficient heating conditions.• More than one-third of households (37%) assess their quality of heating as poor and insufficient.

The percentage is higher (42%) among people over 60. • Households that use electricity for heating, households with below average economic status and

households that heat less than 5 m2 per household member assess their quality of heating as poor and insufficient much more frequently than other households.

• To cut energy expenses, 17% of households reduce the temperature in the rooms they heat, and 27% of households reduce the number of rooms they heat. Households that use electricity for heating and households with below average economic status are more likely than other households to lower the temperature and reduce the number of rooms they heat (see Table 4).

• Households that assess their quality of heating as poor and insufficient, households that heat less than 10 m2 per household member and households whose economic status is below average are more likely to experience health problems (see Table 4).

Table 4. Socioeconomic indicators, by heated space and type of fuel (%, except where otherwise indicated)

Indicator Heated space Type of fuel

less than 10 m2 of heated space per household member

more than 20 m2 of heated space per household member

Electricity district heating

wood and coal

Other All households

IncomeUp to 2500 dinars ($42) per household member in June 2003

33�3 8�4 13�0 4�4 27�3 19�2 20�0

More than 9000 dinars ($150) per household member in June 2003

0�7 21�8 12�3 20�6 2�0 13�7 8

Average total household income per member (US$)

46 102 81 103 54 83 69

Highest educational level attainedPrimary school 38�2 15�4 12�1 8�1 36�3 19�1 26�2University or college degree

7�0 22�8 19�2 25�9 8�3 14�3 13�8

Household size and compositionFive or more members

33�7 9�0 18�9 10�4 26�7 23�8 22�0

Two or more children under the age of 18

24�1 9�1 18�0 9�9 18�6 21�0 16�8

Three- generation household

33�5 12�9 16�7 9�4 30�9 26�0 24�1

Health statusProportion of households without health problems

59�0 69�9 69�2 73�2 60�4 66�5 64�7

Source: perception survey.Note: the differences are even more significant for heated space per member of up to 5 m2. About 47% of households with at least five members and 34% of households with at least two children heat no more than 5 m2 per household member; 40% of these households are three-generation households. About 59% of households that heat no more than 5 m2 had income per household member of less than 2500 dinars ($42) in June, while none had income per household member of more than 9000 dinars ($150).

Annex 2166

INEquAlITIES IN INdOOr AIr POlluTIONThe survey also indicated that in relation to heating conditions, many households suffer from health problems related to indoor air pollution. • Households with at least five members were more likely to use wood or coal for heating (81%).• Households that heat less than 10 m2 per member are most likely to burn wood or coal (81%);

households heating more than 20 m2 per member use district heating (46%), coal and wood (38%), electricity (8%) and gas (5%) (see Table 5).

• Households in apartments with district heating make up a disproportionately high percentage of people with university or college education and people working in trade, transport and communications, finance, public administration, police and military, education, health care and social welfare. Households that burn solid fuels represent a higher than average share of people working in agriculture, trade and crafts.

Table 5. reduction of heated space in apartments, by type of fuel

Space per household member

Electricity district heating wood and coal Other Total

Heated space of less than 10 m2

38�9% 4�2% 51�9% 28�0% 38�5%

Heated space of 10–20 m2

43�5% 44�7% 32�4% 43�0% 37�0%

Heated space more than 20 m2

17�7% 51�0% 15�7% 28�9% 24�5%

All rooms heated 26�7% 95�6% 19�2% 29�1% 37�6%Only some rooms heated

73�3% 4�4% 80�8% 70�9% 62�4%

Average apartment space

72�8 m2 61�9 m2 86�9 m2 109�3 m2* 80�7 m2

Source: perception survey.* This figure reflects the effect of one housing unit of more than 900 m2, which increased the average.

INEquAlITIES IN ASSOCIATEd HEAlTH CONdITIONSThe household surveys shed light on the relation between health problems and lack of heated space and indoor pollution.• People who live in homes with a heated area of less than 10 m2 per household member have a higher

incidence of health problems (41%) than people who live in households with larger heated areas per member (30%).

• There are statistically significant differences in the number of household members with health problems between households using wood for heating (39.6%) and those using district heating (26.8%) (see Table 4).

• There are significant differences in the number of health problems between groups using residential heating based on electricity (12.2%), or using air conditioners (18.4%) or central heating (26.3%), in comparison to the group using naphtha burning stoves (56%), solid fuel light cooking stoves (42.1%) and gas heaters (41.9%).

• The difference in the incidence of health problems and type of heating is most apparent in the case of users of electricity masonry stoves (8.7% having health problems) and users of solid fuels masonry stoves (36.7% having health problems).


CONCluSIONSHouseholds are suffering from lack of heat and high levels of indoor air pollution. During the winter many poor households heat only half of their living space in order to save on energy costs. More than 25% of households heat less than 10 m2 per person, considered to be the necessary minimum. About 60% of the population uses wood and lignite coal – the worst polluters – as their major source of energy for heating, domestic hot water and cooking.

Indoor air pollution is considerable and is correlated with chronic illnesses, including respiratory disease. The health and demographic statistics, survey evidence, mortality patterns and anecdotal evidence together lead to a stark conclusion: indoor pollution is having deleterious effects on 1.5 million households in Serbia and Montenegro.

Health problems and winter mortality are closely affiliated to the above factors. Mortality in Serbia and Montenegro during the winter months can be more than 30% higher than the monthly average mortality rate, with poor households disproportionately affected. Flu epidemics, weather conditions, inadequate clothing, insufficient physical activity, bad nutrition, and high-density housing explain part of the differential, but energy-related factors – such as increased indoor air pollution, inadequate home heating and low temperatures in bedrooms – also play a role, often interacting with other factors.

Annex 2168

ANAlySIS ANd PrESENTATION OF ENvIrONmENTAl HEAlTH INEquAlITIES CONCErNING “GrEEN SPACE” IN SCOTlANd

INTrOduCTIONThe term “green space” is used to describe any vegetated land or water within or adjoining an urban area. Thus, it includes:• disused railway lines, rivers and canals (sometimes called “green corridors”);• woods, grassed areas, parks, gardens, playing fields, children’s play areas, cemeteries and allotments;• countryside immediately adjoining a town which people can access from their homes.11

Green space can also describe derelict, vacant and even contaminated land which has the potential to be transformed. Quality green spaces contribute to improving health and well-being for urban communities. Nested within the Scottish Government’s strategic objective for a healthier Scotland is an objective to sustain and improve health, especially in disadvantaged communities. A taskforce charged with bringing forward an action plan to tackle health inequalities in Scotland identified several key roles of green spaces in creating healthy, sustainable communities. • Green spaces provide valuable opportunities for social, physical, environmental and purposeful

activities that enhance the well-being of the whole population and can be particularly beneficial to people struggling with poor mental health.

• Green spaces improve physical health by providing no-cost opportunities for sports, walking, cycling, gardening and conservation work. They increase levels of physical activity and reduce the risk of people having a range of illnesses or poor health – they are particularly valuable in low-income communities.

• Green spaces support children’s physical and social development by offering opportunities for play and physical activity that are essential for early years development, as well as for older children and teenagers. In low-income communities where private gardens are few, public green spaces are particularly crucial in supporting children’s development.

• Green spaces offer improved community health through the role of green infrastructure and green networks, which reduce health risks from traffic, air pollution, noise flooding and temperature extremes, as well as providing opportunities for physical and social activity and play.

By contrast, degraded and poor quality green space is considered to limit development of healthy sustainable communities by:• discouraging physical activity • undermining well-being • increasing social isolation • damaging community morale • denying children access to outdoor play• increasing health inequalities.

An understanding of how proximity of, and access to, urban green space aligns with social variables is of considerable policy significance and can inform interventions with potential to bear directly on health, well-being and inequities in health. Organized approaches to securing better green space, distributed

11 Greenspace Scotland (2004)� What is greenspace? Stirling, Greenspace Scotland (http://www�greenspacescotland�org�uk/default�asp?page=26, accessed 26 September 2011)�


more equitably between population groups, are consistent with the Scottish Government’s strategic aspirations to deliver a wealthier and fairer and a healthier Scotland. The right sort of green space, appropriately located in relation to population, is also consistent with the remaining three Scottish Government strategic objectives; namely, to create a safer and stronger, a smarter and a greener Scotland. The collection of appropriate measures of green space aligned to social variables is potentially very valuable in policy terms.

Analysis of accessibility of green space raises a further issue. Technically, measuring proximity can be done at various levels of sophistication: firstly, Euclidian distance can be used and a factor incorporated to divide or multiply the distance to account for network distance; secondly, if required, actual access points can be mapped for some green spaces (such as conventional parks) and then used with network analysis. However, proximity is only one factor that impacts on accessibility and use of green space; perception of crime, functionality and design can also influence use and these can be considered under community consultation.

dATA ANd mETHOdOlOGyOver the last decade the analysis of green space has been significantly improved by the availability of detailed georeferenced socioeconomic and health data. These data can be combined with detailed green space data, usually within a GIS. This has allowed researchers to undertake a more detailed analysis of the issue and also aided the presentation of this topic to policy-makers.

Most of these data come from government or quasi-government sources and the opening up of national mapping organizations’ data sets, and much is now available for interactive mapping on web sites. The example of Scotland (similar data and analyses have also been produced in England and Wales) illustrates the data sets available and resulting analyses have been based on the following elements:• datazones – small area units of 500–1000 people, representing the default unit for the collection of

most government data in Scotland;• the Scottish Index of Multiple Deprivation12 – a range of socioeconomic data at datazone units,

based on 38 indicators available in a range of domains, including income, employment, health, education, skills and training, housing, geographic access and crime;

• unit postcodes – approximately 15 households tied to a unique identifier and a classification of these postcodes on an urban rural scale;

• Ordnance Survey (the mapping agency of the United Kingdom Government) “MasterMap Address Layer 2” (based on the Local Land and Property Gazetteer), which maps every individual building in the country;

• detailed classification of green space by type at the local authority level, as well as some nationally available data sets.13

By spatially integrating different sets of data within a GIS it is possible to get an understanding of the spatial distribution of green space, and the quality and potential impacts on health. An early example of this approach was carried out for the Scottish and Northern Ireland Forum for Environmental Research (SNIFFER), comprising the Scottish Executive, Scottish Environmental Protection Agency, Forestry Commission Scotland and Scottish Natural Heritage.14

12 Scottish Government (2010)� Scottish Index of Multiple Deprivation. Edinburgh, Scottish Government (http://scotland�gov�uk/Topics/Statistics/SIMD, accessed 26 September 2011)�

13 Scottish Government (2008)� Planning Advice Note 65. Edinburgh, Scottish Government (http://www�scotland�gov�uk/Resource/Doc/225179/0060935�pdf, accessed 26 September 2011)�

14 Fairburn J et al� (2005)� Investigating environmental justice in Scotland: links between measures of environmental quality and social deprivation. Edinburgh, SNIFFER�

Annex 2170

rESulTSAll of Scotland has now been mapped in detail at the local authority level for green space access. The early proximity (Euclidean) studies have been improved by the use of analysis of routes to green space; this also allows planners to evaluate how many people will be affected by improvements to green space, as well as improvements to possible routes. For example, by classifying the population into 10 deciles by the Scottish Index of Multiple Deprivation it was possible to show that people in deciles 9 and 10 (the two wealthiest) were twice as likely as those in any other decile to be living near a local nature reserve.

Complementing the quantitative/GIS method has been the use of largely qualitative methods for use in community consultation. Greenspace Scotland, Scottish Natural Heritage and the Institute of Occupational Medicine produced a health impact assessment guide for green space, which is intended to aid the assessment of health and equity impacts of existing and proposed green space.15 It is hoped that the guidance will contribute to a greater recognition of green space’s role in improving health and greater emphasis on the health effects of green space in proposals.

Starting in 2010, the Scottish Household Survey is now publishing data on perceived green space quality across Scotland, by Index of Multiple Deprivation and local authority of residence. An example is provided in Fig. 6.

Fig. 6. Percentage of adults (aged 16+) who agreed there were any ‘safe and pleasant’ parks or green spaces available in the area, by Scottish Index of multiple deprivation quintile, 2009

0

10

20

30

40

50

60

70

80

90

100

1 - mostdeprived

2 3 4 5 - leastdeprived

Area

Per

cent

age

CONCluSIONS ANd OPPOrTuNITIES FOr rEPlICATIONThis approach provides a very strong evidence base for policy-makers at the local level and can be used in particular to bring together local authorities to develop integrated and spatially targeted policy. In fact, Forestry Commission Scotland used just such an approach when making grants available to improve green spaces and to plant new woodlands.

The difficulties for many countries in replicating the approach to green space described in this example at the national level (and in some case even at the local level) must be acknowledged. However, it is undoubtedly the case that such an approach could be replicated by local authorities in some countries. Local authorities will generally have a local gazetteer (usually containing information down to the individual house), data on the location of green space and some socioeconomic data for the population. Such an approach does need a reasonably high level of technical competency, especially in GIS. However, there are several detailed guidance manuals available from the United Kingdom demonstrating how to use this approach.

15 Greenspace Scotland (2008)� Health impact assessment of greenspace – a guide. Stirling, Greenspace Scotland (http://www�greenspacescotland�org�uk/upload/File/Greenspace%20HIA�pdf, accessed 26 September 2011)�


ASSESSmENT OF ENvIrONmENTAl HEAlTH INEquAlITIES IN FINlANd: rESIdENTIAl ExPOSurE TO AmbIENT AIr POlluTION FrOm wOOd COmbuSTION ANd TrAFFIC

INTrOduCTIONThe methods and results below are extracted from the final report of the Finnish PILTTI study.16 The main results will also be published in two scientific articles.17 18 In addition to the material briefly summarized below, the PILTTI study also used a GIS to map the distribution of the two types of exposure over Finland, estimated the health impacts and forecasted the changes in exposure and health impacts to 2020.

The aim of the PILTTI project was to assess the adverse health effects from air pollution by primary fine particulate matter (aerodynamic diameter smaller than 2.5 µm (PM2.5)) in Finland from traffic and domestic wood combustion – how the exposures and risks are distributed in space, time and across the sociodemographic strata. PM2.5 is the air contaminant, and most likely also the environmental contaminant, with the highest burden of disease impact in Finland, Europe and the world. The two sources being assessed are those with the greatest contrast in area covered, and thus the greatest exposure and risk differences. They are also the PM2.5 sources on which national regulation and local actions can exert the highest impacts. They therefore provide a relevant target for environmental health inequality assessment.

dATA ANd mETHOdOlOGyResidential location and sociodemographic population data for Finland were obtained from the Statistics Finland Grid Database (www.stat.fi/tup/ruututietokanta/index_en.html). The data set contained population numbers for Finland on a resolution of 250 x 250 m2 for 2004. For exposure modelling, the population data were transformed into 1 km2 spatial resolution with GIS-based modelling platform ArcMap 9.2.

Source emission data for domestic wood combustion and traffic were divided into two categories because of different temporal variations in the data, and different dispersion characteristics. The four categories were: • domestic wood combustion in residential buildings• domestic wood combustion in recreational buildings• direct traffic emissions, tailpipe, tyre and brake wear• suspended traffic emissions (road dust).

16 Ahtoniemi P et al� (2010)� Health risks from nearby sources of fine particulate matter: domestic wood combustion and road traffic (PILTTI)� Helskinki, National Institute for Health and Welfare�

17 Taimisto P et al. (2011). Evaluation of intake fractions for different subpopulations due to primary fine particulate matter (PM2.5) emitted from domestic wood combustion and traffic in Finland. Air Quality, Atmosphere & Health, 4(3–4):199–209�

18 Karvosenoja N et al. (2011). Integrated modeling assessments of the population exposure in Finland to primary PM2.5 from traffic and domestic wood combustion on the resolutions of 1 and 10 km� Air Quality, Atmosphere & Health, 4(3–4):179–188�

Annex 2172

PM2.5 emissions of domestic wood combustion and road traffic were calculated with the Finnish Regional Emission Scenario model of the Finnish Environment Institute.19 The emissions are calculated from the parameters of activity levels, emission factors and emission control technologies for Finland in 2000 and disaggregated to 1 km and 1 hour resolutions.

Emission–exposure relationships for emissions of domestic wood combustion in residential and recreational buildings, as well as direct and suspended traffic emissions for the entire population of Finland, were estimated using the intake fraction (iF) concept. The estimation of iFs for air pollution emissions are based on emission, dispersion, concentration, population and breathing rate data.

The applied atmospheric dispersion model was the Urban Dispersion Modelling system developed at the Finnish Meteorological Institute. Meteorological data from 10 synoptic weather stations from 2000 to 2003 were used in the calculations.

Incremental PM2.5 concentrations were calculated for incremental unit emissions with a 1 km grid resolution for each of the four source categories – in total for 783 900 grid cells. Each source was assumed to be in the centre of a 40 x 40 km2 grid for domestic wood combustion and 20 x 20 km2 grid for road traffic emissions. Contributions from the more distant sources to the concentration in a grid cell were ignored. The iFs, and further the exposure and risk estimates for the populations and sources in each grid cell, were calculated using the actual source emission and population data with statistical computation and graphic system R version 2.7.0.

rESulTSBecause the high spatial resolution population statistics also include data on the sex, age and education level of each individual, the average exposures to the four PM2.5 categories from the two traffic and two wood combustion sources, and thus the exposure differences between the different sociodemographic groups, could be estimated. These are shown in Fig. 7.

Females appear to experience slightly higher exposures to traffic PM2.5 but equal exposures to wood combustion PM2.5 compared to males. The difference is somewhat unexpected for traffic PM2.5, but is small and statistically insignificant.

Children experience the lowest and adults the highest exposures to traffic particles. The elderly experience the lowest exposures to wood combustion particles. The differences, however, are small, and the presumably higher vulnerability of children may more than eliminate the advantage gained from lower exposure.

Those with high school and higher education experience significantly higher exposures to traffic PM2.5 and also higher exposure to wood combustion PM2.5 than those with lower levels of education. Regarding exposure to traffic PM2.5, the average exposure level is highest for those with high school education, but the college/university educated also experience significantly higher exposure levels than those with vocational school education or less. The differences in the exposures to PM2.5 from wood combustion are small in comparison, but increase consistently with the education level.

Compared to international studies, these findings could be considered unexpected, yet it is quite credible in the context of Finnish cities, where on the one hand the most expensive housing is often found in city centres close to the busiest traffic, and on the other hand wood combustion is a supplementary source of heat in the mostly electrically-heated suburban homes that lie outside of district heating networks. The lower socioeconomic groups, conversely, are more likely to live in suburban high-rise buildings, heated via district heating pipelines, with no equipment for wood burning, and distanced from the busiest

19 Karvosenoja N� (2008)� Emission scenario model for regional air pollution� Helsinki, Finnish Environment Institute (Monographs of the Boreal Environment Research, No. 32; http://lib.tkk.fi/Diss/2008/isbn9789521131851/isbn9789521131851.pdf, accessed 26 September 2011).


traffic zones. The results highlight the fact that the impacts of many, if not most, SES parameters on pollution exposures are not universal, but are affected by local social, economic and urban development histories and cultures, and therefore require local analysis with local data.

Note that the exposure data were estimated from outdoor concentrations; the differences in the outdoor–indoor transportation of particles were ignored, the data on exposure to traffic particles do not include personal exposure while in transport, and the data on exposure to wood combustion particles do not include indoor exposure to particles emitted from fireplaces directly into indoor air. Taking into consideration the additional more direct fields of traffic and indoor sources would amplify the estimated differences.

Fig. 7. Average exposure (μg/m3) of different subpopulations to primary Pm2.5 from domestic wood combustion and traffic in Finland in 2000

Sex

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Traffic exhaust and wear

Trafficsuspension

Residential buildings Recreational buildings

Ave

rage

exp

osur

e (µ

g/m

³)

Female Male

Age

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Traffic exhaust and wear

Trafficsuspension

Residential buildings Recreational buildings

Ave

rage

exp

osur

e (µ

g/m

³) Children (0–17)Adults (18–64)Elderly (65+)

Education

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

Traffic exhaustand wear

Trafficsuspension

Residentialbuildings

Recreational buildings

Emission source category

Ave

rage

exp

osur

e (µ

g/m

³)

Higher educationHigh schoolVocational educationComprehensive school

Note: the levels are averages of exposure to PM2.5 from specific sources; for example, the average exposure level of the high school-educated population to traffic generated PM2.5 (exhaust, tyre and brake wear and road dust) is 1.7 + 1.0 = 2.7µg/m3. The total average PM2.5 exposure from all sources is approximately 10 µg/m3.

Annex 2174

CONCluSIONSWhile there are continuing national and EU-based policy actions to reduce exposure to traffic and heating generated PM2.5, these results have not led to any actions for reducing the socioeconomic differences in exposure and risk. The reasons are the small and insignificant differences between the sex and age groups, the relatively low exposure of children, and the fact that the higher exposure of the more educated does not reflect social deprivation or lack of choice and is compensated by their generally better underlying health status. However, the data have been useful to confirm that no action is required due to a very low level of environmental health inequalities in the Finnish population.

The analysis method is in itself universally applicable, but requires individual level residential location, housing and sociodemographic data, which are available from the census data in Finland but not in all countries. It also requires data on wood heating in each building and traffic flows on each road link, which are prohibitively expensive to generate if not otherwise available.


ENvIrONmENTAl HEAlTH INEquAlITy rEPOrT ON THE ImPACT OF SES ON THE PrEvAlENCE OF AllErGIES ANd rESPIrATOry dISEASES ANd SymPTOmS IN HuNGArIAN CHIldrEN

INTrOduCTIONThe aim of the National Children’s Respiratory Health Survey was to assess the relationship between the prevalence of allergic and asthmatic symptoms and chronic bronchitis in third grade schoolchildren and social deprivation at the settlement level in Hungary. This representative study is the first attempt to assess the impact of social inequalities on respiratory health in Hungarian children at a community (settlement) level.

dATA ANd mETHOdOlOGyA national cross-sectional study was performed in autumn 2005.20 All children in third grade classes with more than 10 children in any elementary school of Hungary were invited to participate. Questionnaires to be completed by the parents were sent to the head teachers of the schools, accompanied by a letter about the objectives of the study. The questionnaires served as sources of information on the children’s present and past health status, perinatal conditions, the parents’ respiratory health and smoking habits, the home environment and the SES of the family.

2726 schools were invited and 2160 schools (79.2%) participated in the survey. Out of the 82 082 questionnaires sent out to these schools, 62 711 were returned (76.4%). Based on the questionnaire information, combined data on variables including chronic bronchitic symptoms, asthmatic symptoms during the last 12 months and allergic symptoms were constructed. The presence of chronic bronchitic symptoms was based on at least one positive answer to three questions (whether the child usually coughs in the morning, during the day or at night in the autumn/winter season; whether the child coughed on most days for at least three months consecutively in the last autumn/winter season; whether the child usually coughs up phlegm when he/she does not have a cold). The assessment of asthmatic symptoms was based on at least one positive answer to five questions, all related to the last 12 months (whether the child’s chest sounded wheezy or whistling; whether the child’s chest sounded wheezy during or after exercise; whether the child was woken up by wheezing; whether the child suffered from a dry cough at night; whether the child was treated for asthma). There was a separate question about whether asthma had ever been diagnosed by a physician. Questions on allergies covered allergies to house dust, pets, pollen, mould, food, medicine or anything else, and whether an allergy had been diagnosed by a doctor.

20 Paldy A et al� (2009)� Impact of socioeconomic state on the prevalence of allergic and respiratory symptoms and diseases and in Hungarian children population� Epidemiology, 20(6):S178� The study was funded by European health and environment information system for risk assessment and disease mapping (EUROHEIS) project no� 2006126�

Annex 2176

CHArACTErIzATION OF SES OF THE POPulATION AT THE SETTlEmENT lEvEl: dEvElOPmENT OF A dEPrIvATION INdExTo develop the deprivation index, socioeconomic indicators were chosen at the settlement level from the Ministry of Local Government and Regional Development’s regional informational system database. The data were derived originally from the Hungarian Central Statistical Office (2001 census) and the Hungarian Tax and Financial Control Administration. The selected indicators were: • income (gross income serving as a basis of personal income tax per permanent head of population); • low qualification level (proportion of the total population older than 15 years with only basic

education or illiterate); • unemployment (proportion of total population of working age); • single-parent families (proportion of families with children); • large families (proportion of families with three or more children); • density of housing (number of people per room); • car ownership (passenger cars per 100 inhabitants).

Fig. 8. Spatial distribution of SES in the settlements, banded by quartile

The variables were transformed using natural log transformation and standardization. Each variable was standardized as a z-score based on individual observations, the arithmetic mean of all observations, and the standard deviation. The weight of each indicator was determined from the statistical relationships between them. Individual weights were estimated using factor analysis. The area-specific index was calculated as a weighted sum of the z-scores, with higher values representing greater deprivation: a positive index indicates that the SES is lower than the average, and the converse was shown by negative index values.21

The descriptive ecological study was carried out using Rapid Inquiry Facility (RIF) software and the spatial inequality of morbidity was defined by the RIF “disease mapping” tool, expressed as relative risks obtained by using the empirical Bayes method. Investigation of the association between deprivation and

21 Juhász A et al� (2010)� Development of a Deprivation Index and its relation to premature mortality due to diseases of the circulatory system in Hungary, 1998–2004� Social Science & Medicine, 70(9):1342–9�


the spatial distribution of morbidity from diseases of the respiratory system in Hungary was carried out using the RIF “risk analysis” tool. In the risk analysis module the settlements were grouped into deprivation quartiles, and morbidity rates and risks were calculated for each quartile. Chi-square tests for homogeneity and for linear trend were implemented to test global associations of the deprivation index and relative risk of morbidity from diseases of the respiratory system.

rESulTS

distribution of socially disadvantaged areasTwo large deprived areas were identified in the north-eastern/eastern and south-western sections of Hungary. Parts of counties Borsod-Abaúj-Zemplén and Szabolcs-Szatmár-Bereg in the north-east proved to be the most deprived and underdeveloped parts of the country (see Fig. 8).

distribution of respiratory diseasesBudapest proved to be the have the lowest risk of respiratory disease in Hungary. High-risk areas were observed in the north-east for chronic bronchitis (see Fig. 9) and asthmatic symptoms in both sexes, and in the central Hungarian region for allergies.

Fig. 9. Territorial inequalities of acute bronchitis symptoms in the settlements

Combination of social statistics and disease riskAs a result of ecological risk analysis, a significant association was detected between the SES quartiles and the relative risk of the diseases and symptoms investigated. An upward trend was observed in the risk of morbidity from diseases of the respiratory system by the degree of deprivation (see Fig. 10).

This study indicated that allergic and asthmatic symptoms in children, as well as symptoms of bronchitis, are more frequent in socially deprived areas of the country. The findings also contradict previous evidence that allergic diseases are more common among wealthier populations.

Annex 2178

Fig. 10. The impact of SES on acute bronchitis

0.55

0.67

0.82

1

1.22

1.49

1.82

2.23

-2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5

SES

Rel

ativ

e ris

k of

acu

te b

ronc

hitis

sym

ptom

s Males

Females

CONCluSIONSThe data show that low levels of education, unemployment, low income and high rates of exposure to passive smoking (supported by the questionnaire data) significantly contribute to the respiratory morbidity of children.

Based on the results of this and other similar inequality-driven surveys, specific and regionally-targeted measures should be taken to decrease social inequalities in Hungary in order to lower the risk of respiratory diseases in deprived areas of the country. The regional “closing-up” programmes of the National Development Plan should also tackle this issue. Health promotion campaigns should focus on raising parental awareness, drawing attention to the impacts of passive smoking.

3

ASSESSMENT OF PRIORITY AREAS FOR

NATIONAL ACTION

ANNEX

Annex 3. Country ranking of priority areas for national action180

AN

NEx

3. A

SSES

SmEN

T O

F Pr

IOr

ITy

Ar

EAS

FOr

N

ATIO

NA

l A

CTI

ON

Indi

cato

rIn

adeq

uate

w

ater

sup

ply

Lack

of fl

ush

toile

t in

dwel

ling

lack

of b

ath

or s

how

er in

dw

ellin

g O

verc

row

ding

d

ampn

ess

in th

e ho

me

Inab

ility

to k

eep

hom

e w

arm

in

win

ter

Inab

ility

to

keep

hom

e co

ol in

su

mm

erFi

g� 3

Fig�

5Fi

g� 6

Fig�

9Fi

g� 1

0Fi

g� 1

1Fi

g� 1

2Fi

g� 1

3Fi

g� 1

4Fi

g� 1

5Fi

g� 1

7Fi

g� 1

9In

equa

lity

dim

ensi

onBy

urb

an/ru

ral

popu

latio

nBy

rela

tive

pove

rty le

vel

By h

ouse

hold

ty

peBy

inco

me

quin

tile

By h

ouse

hold

ty

peBy

hou

seho

ld

type

By in

com

e qu

intil

e By

hou

seho

ld

type

By in

com

e qu

intil

e By

rela

tive

pove

rty le

vel

By h

ouse

hold

ty

pe

By in

com

e qu

intil

e C

omm

ents

on

calc

ulat

ion

of in

equa

lity

rank

ing

Impo

ssib

le

due

to la

rge

num

ber o

f co

untri

es w

ith

full

cove

rage

Ineq

ualit

y ra

tio: i

ncom

e be

low

/abo

ve

rela

tive

pove

rty

thre

shol

d

Ineq

ualit

y ra

tio: s

ingl

e-pa

rent

ho

useh

olds

/all

hous

ehol

ds

Ineq

ualit

y ra

tio: l

owes

t/hi

ghes

t in

com

e qu

intil

e

Ineq

ualit

y ra

tio: s

ingl

e-pa

rent

ho

useh

olds

/al

l hou

seho

lds

with

chi

ldre

n

Ineq

ualit

y ra

tio: s

ingl

e-pa

rent

ho

useh

old/

all

hous

ehol

ds

Ineq

ualit

y ra

tio: l

owes

t/hi

ghes

t in

com

e qu

intil

e

Ineq

ualit

y ra

tio: s

ingl

e-pa

rent

ho

useh

old/

all

hous

ehol

ds

Ineq

ualit

y ra

tio: l

owes

t/hi

ghes

t in

com

e qu

intil

e

Ineq

ualit

y ra

tio: i

ncom

e be

low

/abo

ve

rela

tive

pove

rty

thre

shol

d

Ineq

ualit

y ra

tio: h

ighe

st/

low

est

prev

alen

ce

rate

of t

hree

ho

useh

old

type

s

Ineq

ualit

y ra

tio: l

owes

t/hi

ghes

t in

com

e qu

intil

e

dis

adva

ntag

ed g

roup

Rura

l are

asBe

low

rela

tive

pove

rty le

vel

(all

repo

rting

co

untri

es)

Bala

nced

: si

ngle

-par

ent

hous

ehol

ds

in 1

4, a

ll ho

useh

olds

in

16 c

ount

ries

Low

est i

ncom

e qu

intil

e (2

9 of

30

cou

ntrie

s)

Sing

le-p

aren

t ho

useh

olds

(2

4 of

30

coun

tries

)

Sing

le-p

aren

t ho

useh

olds

(2

9 of

30

coun

tries

)

Low

est i

ncom

e qu

intil

e (a

ll re

porti

ng

coun

tries

)

Sing

le-p

aren

t ho

useh

olds

(2

8 of

30

coun

tries

)

Low

est i

ncom

e qu

intil

e (a

ll re

porti

ng

coun

tries

)

Belo

w re

lativ

e po

verty

leve

l (a

ll re

porti

ng

coun

tries

)

Sing

le-p

aren

t ho

useh

olds

(2

3 of

30

coun

tries

)

Low

est i

ncom

e qu

intil

e (2

5 of

27

cou

ntrie

s)

Alb

ania

And

orra

Arm

enia

Aus

tria

*A

zerb

aija

nbe

laru

s be

lgiu

mbo

snia

and

Her

zego

vina

bulg

aria

**

Cro

atia

Cyp

rus

*

*C

zech

rep

ublic

*d

enm

ark

Esto

nia

*Fi

nlan

dFr

ance

*

HO

uSI

NG

-rEl

ATEd

INEq

uA

lITI

ES


Geo

rgia

Ger

man

yG

reec

e*

**

Hun

gary

**

*Ic

elan

d*

Irel

and

*Is

rael

Ital

y*

Kaza

khst

anKy

rgyz

stan

latv

ia*

lith

uani

a*

luxe

mbo

urg

*m

alta

mKd

[a]

mon

aco

mon

tene

gro

Net

herl

ands

Nor

way

*Po

land

Port

ugal

*re

publ

ic o

f mol

dova

rom

ania

*ru

ssia

n Fe

dera

tion

San

mar

ino

Serb

iaSl

ovak

ia*

*Sl

oven

iaSp

ain

Swed

en*

Switz

erla

ndTa

jikis

tan

Turk

eyTu

rkm

enis

tan

ukr

aine

uni

ted

King

dom

**

uzb

ekis

tan

lege

nd:

For d

etai

ls, p

leas

e se

e m

etho

d ex

ampl

e in

C

hapt

er 6

�

Cou

ntri

es w

ith s

ugge

sted

pri

ority

for

natio

nal a

ctio

n (d

efine

d as

cou

ntri

es b

eing

in th

e fo

urth

qua

rtile

for

both

the

abso

lute

and

the

rela

tive

ineq

ualit

y di

men

sion

s, o

r in

th

e th

ird

quar

tile

for

one

and

in th

e fo

urth

qua

rtile

for

the

othe

r in

equa

lity

dim

ensi

on)

Cou

ntri

es th

at w

ould

be

clas

sifie

d as

hav

ing

a su

gges

ted

prio

rity

for

natio

nal a

ctio

n if

they

fell

into

a h

ighe

r qu

artil

e fo

r ju

st o

ne o

f the

two

ineq

ualit

y di

men

sion

sC

ount

ries

with

no

sugg

este

d pr

iori

ty fo

r na

tiona

l act

ion

mis

sing

dat

aN

otes

: * In

equa

lity

ratio

reve

rsed

; [a]

ISO

cod

e fo

r the

form

er Y

ugos

lav

Repu

blic

of M

aced

onia

�


Indi

cato

rw

ork-

rela

ted

inju

ries

Fata

l tra

nspo

rt-r

elat

ed in

juri

esu

nint

entio

nal f

atal

poi

soni

ngFa

tal f

alls

Fig�

21

Fig�

23

Fig�

25b

Fig�

25a

Fig�

26

Fig�

29

Tabl

e 6

Tabl

e 6

Fig�

32

Ineq

ualit

y di

men

sion

By s

exBy

age

gro

upM

orta

lity

rate

of a

ll tra

nspo

rt in

jurie

s by

age

gro

up

RTI m

orta

lity

rate

by

age

gro

upRT

I mor

talit

y ra

te

by s

exPo

ison

ing

mor

talit

y ra

te (a

ll ca

uses

) by

sex

Pois

onin

g m

orta

lity

rate

(exc

ludi

ng

alco

hol p

oiso

ning

) by

sex

Alc

ohol

poi

soni

ng

mor

talit

y ra

te b

y se

x

Fall

mor

talit

y ra

te

by s

ex

Com

men

ts o

n ca

lcul

atio

n of

in

equa

lity

rank

ing

Ineq

ualit

y ra

tio:

mal

e/fe

mal

eIn

equa

lity

ratio

: hi

ghes

t/low

est

inci

denc

e of

thre

e ag

e gr

oups

Ineq

ualit

y ra

tio:

high

est/l

owes

t m

orta

lity

rate

of

four

age

gro

ups

Ineq

ualit

y ra

tio:

high

est/l

owes

t m

orta

lity

rate

of

four

age

gro

ups

Ineq

ualit

y ra

tio:

mal

e/fe

mal

eIn

equa

lity

ratio

: m

ale/

fem

ale

Ineq

ualit

y ra

tio:

mal

e/fe

mal

eIn

equa

lity

ratio

: m

ale/

fem

ale

Ineq

ualit

y ra

tio:

mal

e/fe

mal

e

dis

adva

ntag

ed g

roup

Mal

es (a

ll re

porti

ng

coun

tries

)Va

ries

stro

ngly

Varie

s st

rong

lyVa

ries

stro

ngly

Mal

es (a

ll re

porti

ng

coun

tries

)M

ales

(all

repo

rting

co

untri

es)

Mal

es (a

ll re

porti

ng

coun

tries

)M

ales

(27

of 2

8 co

untri

es)

Mal

es (a

ll re

porti

ng

coun

tries

)

Alb

ania

And

orra

Arm

enia

Aus

tria

Aze

rbai

jan

bela

rus

belg

ium

bosn

ia a

nd H

erze

govi

na

bulg

aria

Cro

atia

Cyp

rus

Cze

ch r

epub

lic

den

mar

k

Esto

nia

Finl

and

Fran

ce

Geo

rgia

Ger

man

y

Gre

ece

Hun

gary

Icel

and

Irel

and

Isra

el

Ital

y

Kaza

khst

an

INju

ry-r

ElAT

Ed IN

Equ

AlI

TIES


Kyrg

yzst

an

latv

ia

lith

uani

a

luxe

mbo

urg

mal

ta

mKd

[a]

mon

aco

mon

tene

gro

Net

herl

ands

*

Nor

way

Pola

nd

Port

ugal

repu

blic

of m

oldo

va

rom

ania

russ

ian

Fede

ratio

n

San

mar

ino

Serb

ia

Slov

akia

Slov

enia

Spai

n

Swed

en

Switz

erla

nd

Tajik

ista

n

Turk

ey

Turk

men

ista

n

ukr

aine

uni

ted

King

dom

[b]

[b]

uzb

ekis

tan

lege

nd:

For d

etai

ls, p

leas

e se

e m

etho

d ex

ampl

e in

Cha

pter

6�

Cou

ntri

es w

ith s

ugge

sted

pri

ority

for

natio

nal a

ctio

n (d

efine

d as

cou

ntri

es b

eing

in th

e fo

urth

qua

rtile

for

both

the

abso

lute

and

the

rela

tive

ineq

ualit

y di

men

sion

s, o

r in

the

thir

d qu

artil

e fo

r on

e an

d in

the

four

th q

uart

ile fo

r th

e ot

her

ineq

ualit

y di

men

sion

)C

ount

ries

that

wou

ld b

e cl

assi

fied

as h

avin

g a

sugg

este

d pr

iori

ty fo

r na

tiona

l act

ion

if th

ey fe

ll in

to a

hig

her

quar

tile

for

just

one

of t

he tw

o in

equa

lity

dim

ensi

ons

Cou

ntri

es w

ith n

o su

gges

ted

prio

rity

for

natio

nal a

ctio

nm

issi

ng d

ata

Not

es: *

Ineq

ualit

y ra

tio re

vers

ed; [

a] IS

O c

ode

for t

he fo

rmer

Yug

osla

v Re

publ

ic o

f Mac

edon

ia; [

b] d

ata

for w

ork

inju

ries

are

base

d on

val

ues

for G

reat

Brit

ain

as U

K da

ta a

re n

ot a

vaila

ble�


Indi

cato

rN

oise

exp

osur

e at

hom

ela

ck o

f acc

ess

to r

ecre

atio

nal o

r gr

een

area

sSe

cond

-han

d sm

oke

expo

sure

at h

ome

Seco

nd-h

and

smok

e ex

posu

re a

t wor

k

Fig�

34

Fig�

36

Fig�

36

Fig�

37

Fig�

39

Fig�

39

Fig�

40

Fig�

42

Fig�

42

Fig�

43

Ineq

ualit

y di

men

sion

By re

lativ

e po

verty

leve

lBy

sex

By in

com

e qu

artil

e By

diffi

culty

pa

ying

bill

sBy

sex

By s

elf-a

sses

sed

soci

al p

ositi

on

By d

ifficu

lty

payi

ng b

ills

By s

exBy

sel

f-ass

esse

d so

cial

pos

ition

B

y oc

cupa

tion

Com

men

ts o

n ca

lcul

atio

n of

in

equa

lity

rank

ing

Ineq

ualit

y ra

tio:

inco

me

belo

w/

abov

e re

lativ

e po

verty

thre

shol

d

Ineq

ualit

y ra

tio:

fem

ale/

mal

eIn

equa

lity

ratio

: lo

wes

t/hig

hest

in

com

e qu

artil

e

Ineq

ualit

y ra

tio:

diffi

culty

pay

ing

bills

mos

t of t

ime/

no d

ifficu

lty

Ineq

ualit

y ra

tio:

fem

ale/

mal

eIn

equa

lity

ratio

: lo

w/h

igh

soci

al

posi

tion

Ineq

ualit

y ra

tio:

diffi

culty

pay

ing

bills

mos

t of t

ime/

no d

ifficu

lty

Ineq

ualit

y ra

tio:

mal

e/fe

mal

eIn

equa

lity

ratio

: lo

w/h

igh

soci

al

posi

tion

Ineq

ualit

y ra

tio:

man

ual w

orke

r/m

anag

er

dis

adva

ntag

ed g

roup

Belo

w re

lativ

e po

verty

leve

l (24

of

30

coun

tries

)

Fem

ales

(19

of 3

1 co

untri

es)

Hig

hest

inco

me

quar

tile

(18

of 3

1 co

untri

es)

Diffi

culty

pay

ing

bills

mos

t of

time

(23

of 3

1 co

untri

es)

Bala

nced

: fe

mal

es in

16,

m

ales

in 1

4 co

untri

es

Low

soc

ial

posi

tion

(26

of 3

0 co

untri

es)

Diffi

culty

pay

ing

bills

mos

t of

time

(24

of 2

8 co

untri

es)

Mal

es (2

5 of

30

coun

tries

)Lo

w s

ocia

l po

sitio

n (2

1 of

28

coun

tries

)

Man

ual w

orke

rs

(25

of 3

0 co

untri

es)

Alb

ania

And

orra

Arm

enia

Aus

tria

**

**

*

Aze

rbai

jan

bela

rus

belg

ium

*

bosn

ia a

nd H

erze

govi

na

bulg

aria

**

Cro

atia

**

**

Cyp

rus

**

**

Cze

ch r

epub

lic*

den

mar

k*

Esto

nia

**

*

Finl

and

**

Fran

ce*

*

Geo

rgia

Ger

man

y*

Gre

ece

**

**

**

Hun

gary

**

**

*

Icel

and

Irel

and

*

Isra

el

Ital

y

Kaza

khst

an

ENv

IrO

Nm

ENT-

rElA

TEd

INEq

uA

lITI

ES


Kyrg

yzst

an

latv

ia*

lith

uani

a*

**

**

luxe

mbo

urg

**

mal

ta*

*

mKd

[a]

**

*

mon

aco

mon

tene

gro

Net

herl

ands

**

Nor

way

*

Pola

nd*

**

**

Port

ugal

**

**

*

repu

blic

of m

oldo

va

rom

ania

**

**

*

russ

ian

Fede

ratio

n

San

mar

ino

Serb

ia

Slov

akia

**

*

Slov

enia

*

Spai

n*

*

Swed

en*

**

Switz

erla

nd

Tajik

ista

n

Turk

ey*

**

*

Turk

men

ista

n

ukr

aine

uni

ted

King

dom

*

uzb

ekis

tan

lege

nd:

For d

etai

ls, p

leas

e se

e m

etho

d ex

ampl

e in

Cha

pter

6�

Cou

ntri

es w

ith s

ugge

sted

pri

ority

for

natio

nal a

ctio

n (d

efine

d as

cou

ntri

es b

eing

in th

e fo

urth

qua

rtile

for

both

the

abso

lute

and

the

rela

tive

ineq

ualit

y di

men

sion

s, o

r in

the

thir

d qu

artil

e fo

r on

e an

d in

the

four

th q

uart

ile fo

r th

e ot

her

ineq

ualit

y di

men

sion

)C

ount

ries

that

wou

ld b

e cl

assi

fied

as h

avin

g a

sugg

este

d pr

iori

ty fo

r na

tiona

l act

ion

if th

ey fe

ll in

to a

hig

her

quar

tile

for

just

one

of t

he tw

o in

equa

lity

dim

ensi

ons

Cou

ntri

es w

ith n

o su

gges

ted

prio

rity

for

natio

nal a

ctio

nm

issi

ng d

ata

Not

es: *

Ineq

ualit

y ra

tio re

vers

ed; [

a] IS

O c

ode

for t

he fo

rmer

Yug

osla

v Re

publ

ic o

f Mac

edon

ia�

4

COUNTRY ABBREVIATIONS

ANNEX

Annex 4. Country abbreviations188

Alb Albania

ANd Andorra

Arm Armenia

AuT Austria

AzE Azerbaijan

blr Belarus

bEl Belgium

bIH Bosnia and Herzegovina

bul Bulgaria

CrO Croatia

CyP Cyprus

CzH Czech Republic

dEN Denmark

EST Estonia

FIN Finland

FrA France

GEO Georgia

dEu Germany

GrE Greece

HuN Hungary

ICE Iceland

IrE Ireland

ISr Israel

ITA Italy

KAz Kazakhstan

KGz Kyrgyzstan

lvA Latvia

lTu Lithuania

lux Luxembourg

mAT Malta

mON Monaco

mNE Montenegro

NET Netherlands

NOr Norway

POl Poland

POr Portugal

mdA Republic of Moldova

rOm Romania

ruS Russian Federation

Smr San Marino

Srb Serbia

SvK Slovakia

SvN Slovenia

SPA Spain

SwE Sweden

SwI Switzerland

TjK Tajikistan

mKd The former Yugoslav Republic of Macedonia (ISO code)

Tur Turkey

TKm Turkmenistan

uKr Ukraine

uNK United Kingdom

uzb Uzbekistan

ANNEx 4. COuNTry AbbrEvIATIONS

5

ABBREVIATIONS

ANNEX

Annex 5. Abbreviations190

ANNEx 5. AbbrEvIATIONS

APAT Environmental Protection Agency (of Italy)

CO carbon monoxideCSdH Commission on Social Determinants

of HealthdAly disability-adjusted life yeardb(A) A-weighted decibelsdmdb European detailed mortality databaseEHESP School of Public Health (of France)ENHIS Environment and Health

Information SystemEqlS European Quality of Life SurveyEu European UnionEu-SIlC EU Statistics on Income and Living

ConditionsEu15 the 15 countries belonging to the EU

before May 2004Eu27 the 27 countries belonging to the EU

after January 2007Ewd excess winter deathFEANTSA European Federation of National

Organisations working with the Homeless

Gbd global burden of diseaseGerES Iv German Environmental Survey

2003/6 for ChildrenGIS geographical information systemHiAP Health in all policiesICd-10 WHO International Statistical

Classification of Diseases and Related Health Problems, tenth revision

iF intake fractionINErIS National Institute for Industrial

Environment and Risks (of France)INSEE National Institute of Statistics and

Economic Studies (of France)ISO International Organization for

Standardization

jmP Joint Monitoring Programme (for Water Supply and Sanitation)

KiGGS German Health Survey for Children and Adolescents

mdb European mortality databasemKd ISO code for the former Yugoslav

Republic of MacedoniamTl1 mortality tabulation list 1NEHAP National Environmental Health

Action PlanNmS12 the 12 new Member States which

joined the European Union in 2004 and 2007

NuTS Nomenclature of Units for Territorial Statistics

OECd Organisation for Economic Co-operation and Development

OHCHr Office of the High Commissioner for Human Rights

Pm2.5particulate matter with an aerodynamic diameter smaller than 2.5 µm

PrTr Pollutant Release and Transfer Register

PTSd post-traumatic stress disorderrIF Rapid Inquiry Facility (software)rTI road traffic injurySES socioeconomic statusSlC Survey of Living Conditions (of

Norway)SNIFFEr Scottish and Northern Ireland

Forum for Environmental ResearchSr sex ratiouNdP United Nations Development

ProgrammeuNECE United Nations Economic

Commission for Europe

10

20

30

40

50

Envi

ronm

enta

l hea

lth

ineq

ualit

ies

in E

urop

e

Environm

ental health inequalities in Europe

Recent debate on the social determinants of health has indicated that the unequal distribution of health and well-being in national populations is a major challenge for public health governance. �is is equally true for environmental health condi-tions and for exposure to environmental risk, which varies strongly by a range of sociodemographic determinants and thus causes inequalities in exposure to – and potentially in disease resulting from – environmental conditions.

Interventions tackling such environmental health inequalities need to be based on an assessment of their magnitude and on the identi�cation of population groups that are most exposed or most vulnerable to environmental risks. However, data to quantify the environmental health inequality situation are not abundant, making comprehensive assessments di�cult at both national and international levels. Following up on the commitments made by Member States at the Fifth Ministe-rial Conference on Environment and Health in Parma, Italy (2010), the WHO Regional O�ce for Europe has carried out a baseline assessment of the magnitude of environmental health inequality in the European Region based on a core set of 14 inequality indicators. �e main �ndings of the assessment report indicate that socioeconomic and demographic inequalities in risk exposure are present in all countries and need to be tackled throughout the Region. However, the report also demonstrates that each country has a speci�c portfolio of inequalities, document-ing the need for country-speci�c inequality assessments and tailored interventions on the national priorities.

World Health OrganizationRegional Office for Europe

Scherfigsvej 8, DK-2100 Copenhagen Ø, DenmarkTel.: +45 39 17 17 17. Fax: +45 39 17 18 18E-mail: [email protected] site: www.euro.who.int